<<

MOLECULAR MECHANISMS BY WHICH BINDS TO AND ACTIVATES THE KAPPA-

by

FENG YAN

Dissertation Adviser: Bryan L. Roth, M.D./Ph.D.

Submitted in partial fulfillment of the requirements

for the degree of Doctor of Philosophy

Department of

CASE WESTERN RESERVE UNIVERSITY

May, 2008

CASE WESTERN RESERVE UNIVERSITY

SCHOOL OF GRADUATE STUDIES

We hereby approve the dissertation of

Feng Yan ______candidate for the Ph.D. degree *.

(signed) Vernon E. Anderson, Ph.D. ______(chair of the committee)

Bryan L. Roth, M.D., Ph.D. ______

Martin D. Snider, Ph.D. ______

Paul R. Ernsberger, Ph.D. ______

Krzysztof Palczewski, Ph.D. ______

______

May, 2008 (date) ______

* We also certify that written approval has been obtained for any

proprietary material contained therein.

TABLE OF CONTENTS

List of Tables vi

List of Figures viii

List of Abbreviations ix

Acknowledgements xiv

Abstract xvi

CHAPTER 1: Introduction

1.1 G -Coupled Receptors………………………………………..……………..…1

1.1.1 G Protein-Coupled Receptor Overview…………………………………………1

1.1.2 G Protein-Coupled Receptor Structures………………………………………...1

1.1.3 G Protein-Coupled Receptor Function……………………..…………………...3

1.2 Opioid Receptors…………………………………………….………………………..6

1.2.1 Overview……………………………………………………...6

1.2.2 Opioid Receptor Function………………………………………………………8

1.2.3 Opioid Receptor Signaling Regulation………………………………………...11

1.3 ………………………………………………………………………………11

1.3.1 Endogenous Opioids…………………………..……………………………….11

1.3.2 Exogenous Opioids………………………..…………………………………...13

1.3.2.1 MOR ligands……..……………………………………………..………..18

1.3.2.2 DOR ligands ……..……………………………………………….…..…18

1.3.2.3 KOR ligands ……..……………………………………………….……..19

i

1.3.3 Opioids’ Medical Use and Abuse…………………….……………...... 20

1.4 The History of and Salvinorin A Research……………………….22

CHAPTER 2: Materials and Methods

2.1 Materials……………………………………………………………………………..27

2.1.1 Chemicals………………………………………………………………………27

2.1.1.1 Commercial chemicals……………………..…………………………….27

2.1.1.2 Synthesized salvinorin A derivatives……………...…………….………27

2.1.2 cDNA constructs……………………………………………………………….28

2.1.2.1 Sub-cloning of hKOR into pUniversal-signal…………………………...28

2.1.2.2 Generation of KOR mutants by site-directed mutagenesis……………....28

2.1.3 Cells…………………………………………………………………………....28

2.1.3.1 Transient transfection and cell membrane preparation..…………………28

2.1.3.2 Stable expression of G ………………………………………….29

2.2 Methods……………….……………………………………………………………..29

2.2.1 Radioligand Binding Assays………………………………………………...... 30

2.2.1.1 Competition binding assay………………………………………..……...30

2.2.1.2 Saturation binding assay …………………………………………..…….30

2.2.2 SCAM……………………………………………………………………….....31

2.2.2.1 MTSEA reaction…………………………………………………………31

2.2.2.2 SCAM radioligand binding assay………………………………………..32

2.2.2.3 Determination of second order rate constants…………………………..33

2.2.2.4 protection against the MTSEA reaction…………………………33

ii

2.2.3 Functional Assays..………………………………………………………….....34

2.2.3.1 Intracellular calcium mobilization (Ca2+ flux)…………………..……….34

2.2.3.2 [35S]GTPS binding assay..……………………………………………….35

2.2.4 Computer Modeling……………………………………………………………35

2.2.4.1 KOR modeling for binding site………………………………………….36

2.2.4.2 KOR modeling for SCAM……………………………………………….36

2.2.5 Large Quantity of KOR Expression and Purification………………………….39

2.2.5.1 KOR expression………………………………………………………….39

2.2.5.2 Labeling KOR with RB-64 or RB-48……………………………………39

2.2.5.3 Anti-FLAG M2 affinity purification……….……………………………40

2.2.5.4 Ni2+ affinity purification of KOR………………………….…………….40

2.2.5.5 SDS-PAGE and western blot analysis…………………………………..41

2.2.5.6 SDS-PAGE and Coomassie staining…………………………………….41

2.2.6 Mass Spectrometric Analysis…………………………………………………..42

2.2.7 Prepulse Inhibition Animal Study…………………..………………………….43

CHAPTER 3: The Binding Site of Salvinorin A

3.1 Introduction and Rationale…………………………………………….……………46

3.2 Results…………………………..…………………………………………………..50

3.2.1 Receptor-based Binding Site Study …………………………………………...50

3.2.1.1 Key residues in the binding site were identified by a combined

mutagenesis/computer-modeling method………………………………..50

iii

3.2.1.2 Specific ligand-receptor interactions determine the orientation of

salvinorin A in the binding site…………………………………………..54

3.2.1.3 Molecular modeling to visualize and predict the binding site…………...55

3.2.2 Ligand-based Binding Site Study……………………………………………...57

3.3 Discussion……………………………………………………………………………68

CHAPTER 4: The Conformational Changes of KOR Induced by G

Protein Coupling

4.1 Introduction and Rationale…………………………………………………………...74

4.2 Results ……………………………………………………………………………….78

4.2.1 Examine the Conformational Changes of KOR by SCAM……………………78

4.2.1.1 Characterize Cys mutants in TMs 6 and 7 and EL2 of KOR……………78

4.2.1.2 SCAM elucidates conformational changes induced by G proteins……...79

4.2.2 Kinetics Studies Confirmed that Conformational Changes Occur in KOR……84

4.2.3 Conformational changes induced by G protein-coupling have significant effects

on affinities…………………………………………………………….86

4.2.4 Refining the Binding Site of Salvinorin A …………………………………….87

4.3 Discussion……………………………………………………………………………90

CHAPTER 5: The Design and Application of Covalently-Bound Agonist to Probe KOR

5.1 Introduction and Rationale…………………………………………………………...95

5.2 Results…………………………..…………………………………………………..100

iv

5.2.1 Characterizing Covalently-Bound Ligands by Molecular

Methods………………………………………………………………………100

5.2.1.1 RB-64 and RB-48 are potent KOR ……………………………100

5.2.1.2 Covalent labeling WT KOR or mutants with RB-64…………………...101

5.2.2 Characterizing Covalently-Bound Ligands by ………….104

5.2.2.1 KOR expression, labeling and purification……………………………..104

5.2.2.2 Mass spectrometric analysis……………………………………………105

5.2.3 In vivo PPI Study of RB-64 ……………………...…………………..….…...111

5.3 Discussion…………………………………………………………………………..115

CHAPTER 6: Future Directions and Conclusions

6.1 Future Directions…………………………………………………………………...119

6.1.1 Quantifying G protein Expression……………………………………………119

6.1.2 Chasing the Active and Global Conformation of KOR……………………...119

6.1.3 Crystallographic Study of KOR………………..………………………..……122

6.2 General Conclusions..………………………………………………………………123

Bibliography………………………………………………………………………...124

v

LIST OF TABLES

1.1 Endogenous opioid ………………………………………………………...12

1.2 Select list of endogenous opioid peptides and derivatives…………...…...14

1.3 Pharmacological effects of salvinorin A in select animal studies……………………25

3.1 Key residues in salvinorin A’s binding site …………………………………………48

3.2 Affinity (Ki, nM) of salvinorin A, U69593 and A (1-13) binding to the WT

KOR and mutants transiently expressed in HEK293 T cells……………..….………51

3.3 Agonist (EC50, nM) and relative agonist efficacy (normalized Emax) of

salvinorin A, U69593 and (1-13) for the WT KOR and mutants

transiently expressed in HEK 293 cells ………………………………………...... …53

3.4 Effect of Cys-substitution mutations on salvinorin A and 2-thiosalvinorin B binding

to KORs ………………………………………………………………………...…...55

3.5 Residues within specified heavy (O, N, etc.) distances between the salvinorin A

ligand and the KOR receptor for the proposed binding mode ……………………....56

3.6 Salvinorin A derivatives tested in Dr. Bryan Roth’s lab …………………...……….59

4.1 The Comparison of Common Biochemical and Biophysical Methods Used for GPCR

Conformation Study………………………………………………………………….77

3 4.2 Kd and Bmax values of [ H] binding to the WT KOR and EL2

mutants………………………………………………………………………..….…..79

4.3 Changes in inhibition upon the coupling of G proteins Gα16 and Gαi2………………81

4.4 Second-order rate constants (k, M-1s-1) of MTSEA reaction with Cys mutants of

KOR………………………………………………………………………………….85

vi

4.5 Differential effects of G protein overexpression on agonists (salvinorin A, U69593

and Dynorphin A (1-13)) binding affinities……………………………………….…86

4.6 Affinity (Ki, nM) of salvinorin A binding to WT KOR and mutants………..……….87

5.1 The pharmacological profile of RB-64 and RB-48…………………...…………….101

5.2 Mascot searches of chymotryptic peptides from KOR (47 kDa) and KOR (52

kDa)…….…………………………………………………………………………...106

5.3 Detected KOR peptides from chymotryptic by AB 4800 MALDI

TOF/TOF…………………………………………………………………………...107

5.4 RB-64 modified peptides predicted by in silico digestion ……….…110

5.5 Null activity levels (mAmp displacement) in mice treated with vehicle, salvinorin A,

or RB-64…………………………………………………………………………….111

vii

LIST OF FIGURES

1.1 GPCR signaling……………………………………………………………………….5

1.2 GPCR signaling regulation based on β2AR model..…………………………………6

1.3 Opioid receptors regulate the activity in VTA through the disinhibition

mechanism…………………………………………………………………………...10

1.4 Chemical backbones of morphine-derived molecules.………………………………14

2.1 Nucleophilic reaction between MTSEA and Cys side chain –SH (S-)………………32

3.1 The pharmacophore of salvinorin A…………………………………………………47

3.2 Modeling salvinorin A-KOR interactions……………………………………………57

4.1 Gα subunit overexpression was used to stabilize receptor conformations in the

various active states………………………………………………………………….76

4.2 SCAM pattern ……………………………………………………………………….83

4.3 KOR model based on active rhodopsin structure……………………………………89

4.4 Second-order rate constants (k, M-1s-1) revealed a directional preference and α-helical

periodicity …………………………………………………………………………...94

5.1 The mechanism for RB-64 labeling (covalently-bound to) KOR……………………99

5.2 RB-64 labeling…..………….………………………………………………………103

5.3 Separation of KOR by SDS-PAGE………………………………………………..105

5.4 Sequence coverage of KOR by MALDI TOF/TOF……………………………..….108

5.5 MALDI mass spectra of KOR after chymotrypsine digestion………………..…...110

5.6 PPI responses to salvinorin A and RB-64 by C57BL/6J mice………….…………..115

6.1 SILAC method for quantitative proteomics………………………………………...120

viii

LIST OF ABBREVIATIONS

5-HT2A 5-hydroxytryptamine 2A (-2A) receptor

7TM seven transmembrane

AC adenylate cyclase

ATP adenosine 5'-triphosphate

β2AR β2- receptor

bn Bieman modified Roepstorff and Fohlman nomenclature ion

BSA bovine serum albumin cAMP adenosine 3', 5'- cyclic monophosphate cDNA cloned DNA

CHCA α-cyano-4-

CID collision-induced dissociation cm centimeter (10-2 meter)

CNBr

CNS

Da dalton

DAG diacylglycerol

DEA Enforcement Administration

DMEM Dulbecco's modified essential medium

DNA deoxyribonucleic acid

DOR δ-opioid receptor

DTT 1,4-dithiothreitol

EDTA diamine tetra-acitic acid

ix

EL1 first extracellular loop

EL2 second extracellular loop

EL3 third extracellular loop

ER

ERK extracellular-regulated kinase

FBS fetal bovine serum

FDA Food and Drug Administration

FTICR Fourier transform ion cyclotron resonance mass spectrometer

Gα G protein subunit

Gαs alpha subunit of stimulatory G protein for

Gαi alpha subunit of inhibitory G protein for adenylyl cyclase

Gβ G protein beta subunit

Gγ G protein gamma subunit

GDP guanosine 5'-diphosphate

GEF guanine-nucleotide exchange factor

GFP green fluorescent protein

GPCR G protein-coupled receptor

GRK G protein-coupled receptor kinase

GTP guanosine 5'-triphosphate

HEK human embryonic

HEPES 4- (2-hydroxyethyl)-1-piperazineethanesulfonic acid

HPLC high performance liquid chromatography

HRP horse radish peroxidase

x

IL1 first intracellular loop

IL2 second intracellular loop

IL3 third intracellular loop

INRC International Research Conference

IP3 inositol-1,4,5-trisphosphate

IRES internal ribosome entry site

JNK c-Jun N-terminal kinase kDa kiloDalton

KOR κ-opioid receptor

LSC liquid sintillation counting

LSD lysergic acid diethylamide

MALDI matrix assisted laser desorption/ionization

MAPK mitogen-activated protein kinase

MD molecular dynamics

MOR µ-opioid receptor

MS mass spectrometry

MS/MS tandem mass spectrometry

µg microgram

µl microliter mg milligram ml milliliter mM millimolar

MTS methanethiosulfonate

xi

MTSEA 2-aminoethylmethanethiosulfonate

MW molecular weight

m/z mass to charge ratio

NA

NCBI National Center for Biotechnology Information

NIDA National Institute on Drug Abuse

nM nanomolar

NMDA N-methyl-D-aspartate

ORL1 opioid receptor-like (an orphan receptor)

PAGE polyacrylamide gel electrophoresis

PBS phosphate buffered saline

PCR polymerase chain reaction

PDB protein data bank

PDSP Screening Program

PI phosphoinositide

PI3K phosphatidylinositol 3 kinase

PIP2 phosphatidylinositol biphosphate

PKA

PKC protein kinase C

PLCβ phospholipase Cβ

PMF peptide mass fingerprinting

PTX pertussis toxin

rpm rotations per minute

xii

RFU relative fluorescence units

SA sinapinic acid (3,5-dimethoxy-4-hydroxycinnamic acid)

SCAM substituted cysteine accessibility method

S. divinorum Salvia divinorum

SDS sodium dodecylsulfate

S/N signal to noise ratio

TBS Tris-buffered saline

TBST Tris-buffered saline with Tween-20

TFA trifluoroacetic acid

TM transmembrane domain

TOF time of flight vdW van der Waals’

VTA

WT wild type yn Bieman modified Roepstorff and Fohlman nomenclature peptide ion

xiii

ACKNOWLEDGEMENTS

I am deeply grateful to my advisor, Dr. Bryan L. Roth, for his outstanding guidance and creative ideas. It has been a great and honor to work in Dr. Bryan L. Roth’s lab since 2003. I have learned not just the art of scientific research, but also the life a scientist should live up to.

I sincerely thank all the members of my dissertation committee. This includes my dissertation committee Chair, Dr. Vernon E. Anderson, as well as Drs. Martin D. Snider,

Paul R. Ernsberger, Krzysztof Palczewski and my advisor Dr. Bryan L. Roth. I am so fortunate to have such excellent guidance and support from them during my doctoral program and my research has benefited greatly by their suggestions and critiques.

I deeply appreciate the support from my fellow graduate students in Dr. Roth’s lab. Ryan

Strachan and Atheir Abbas have been helpful in many ways. I am grateful to them for sharing their talents, research experiences and friendship during the past four and half years. While we may end up in different places, I will always consider them as friends.

I am thankful to my past and present colleagues, including former graduate student

Douglas Sheffler, former postdocs Drs. Timothy A. Vortherms and Blaine N.

Armbruster, present postdocs Drs. Xiping Huang, Niels N. Jensen, John A. Alan and

Ying Pei. They all helped and inspired me in different ways. I thank them for being supportive and understanding.

xiv

To my Mom, Dad and younger brother — Wei, I can’t express in words what your unconditional love and support meant to me. Without you, I couldn’t have come this far.

At last, I would thank all the nice and caring people I have met during the past eight years living in Cleveland, Ohio and Chapel Hill, North Carolina. Your and souls make this world beautiful. Wish you all the best.

xv

Molecular Mechanisms by which Salvinorin A Binds to and Activates the κ-Opioid Receptor

Abstract

by

FENG YAN

Salvinorin A, the most potent naturally-occurring , has gained great attention

since the κ-opioid receptor (KOR) was identified as its principal molecular target (1).

However, the molecular mechanisms by which salvinorin A — a small-molecule agonist

— binds to and activates KOR was unclear. To understand these mechanisms, three aims

were proposed for my dissertation research; correspondingly, I will report our findings in

three parts (Chapter 3, Chapter 4 and Chapter 5) in this dissertation.

The primary goal (Chapter 3) is to identify the binding site of salvinorin A in KOR. A

combination of site-directed mutagenesis and molecular modeling was applied to

determine the structural features of KOR essential for the binding of Salvinorin A (2).

Meanwhile, a series of naturally-occurring and synthetic salvinorin A derivatives was designed and assayed to compare their binding and functional properties (3-6).

The subsequent goal (Chapter 4) is to investigate KOR’s conformational change during the activation process. In this part of the dissertation research, over-expression of Gα16 and Gαi2 were used to increase the coupling ratio between KOR and the Gα subunits (7).

The substituted cysteine accessibility method (SCAM), utilizing the specific reaction

between the thiolate groups (-S-) and 2-aminoethylmethanethiosulfonate (MTSEA), was

xvi

applied to detect the conformational changes of the receptor (7). Intriguingly, these G protein-dependent conformational changes significantly increased the binding affinity of salvinorin A.

In PART III (Chapter 5), our goal is to further verify ligand-receptor interactions by designing a series of ligands capable of covalently binding to KOR. From our earlier work using the SCAM approach, we demonstrated that C3157.38 was both water

accessible and highly reactive to methanethiosulfonate (MTS) reagents (7). Thus far, two

compounds — RB-48 and RB-64 (both with pM potency and extraordinary selectivity for

KOR) — have emerged as being suitable for affinity-labeling KOR. Our preliminary

mass spectrometry data was consistent with C3157.38 as the labeling site.

Collectively, this research project has revealed the molecular mechanisms by which a

small-molecule agonist selectively binds to and activates a Class A GPCR.

xvii

CHAPTER 1: Introduction

1.1 G Protein-Coupled Receptors

1.1.1 G Protein-Coupled Receptor Overview

G protein-coupled receptors (GPCRs) are membrane-spanning proteins that can detect extracellular stimuli and initiate intracellular responses. Human GPCRs are the largest

receptor superfamily with roughly 1000 members, which are grouped into three major

classes according to the authoritative list by the International Committee of

Pharmacology Committee on Receptor Nomenclature and Classification (NC-IUPHAR)

(8). This list is updated constantly on the IUPHAR website (www.iuphar-db.org). The

representative receptors for these three major classes are rhodopsin (Class A), the

receptor (Class B), and the metabotropic glutamate receptors (Class C). The rhodopsin-

like Class A is the largest and currently most studied. GPCRs can recognize highly

distinct stimuli or ligands, including photons, odorants, small-molecules, peptides and

proteins (9, 10). Consequently, GPCRs represent the largest known human -family

and account for 4-5% of the human (11-13). Nearly 30% of the marketed prescription target GPCRs (14). To understand how GPCRs transduce signals has been a major goal in life science for the last few decades (15).

1.1.2 G Protein-Coupled Receptor Structures

The defining structural feature of GPCRs is their seven transmembrane spanning domains

(7TM) linked by three intracellular and three extracellular loops, which forms a helical

1 bundle. Therefore GPCRs are also referred to as 7TM or heptahelical receptors. This

7TM structure of GPCRs was first predicted by Unwin and Henderson in 1975 based on electron diffraction data obtained from bacteriorhodopsin (16). In 2000, this 7TM structure was validated by the first high-resolution crystal structure of a mammalian

GPCR (bovine rhodopsin, resolution 2.80 Å, 1f88) (17, 18). In this structure, rhodopsin is tightly locked into the inactive state by the covalently-bound ligand, 11-cis-retinal. This inactive-state structure of rhodopsin provides a critical template for GPCR homology modeling and has led to a deeper understanding of all Class A GPCRs (10, 19-22). More recently, the structure of rhodopsin in the photoactivated intermediate state was also achieved (23). In 2007, the first high-resolution structure of a human GPCR (β2- adrenergic receptor (β2AR)-T4 lysozyme fusion protein, resolution 2.40 Å, 2rh1) was solved (24-27). To restrain the flexible internal loops of β2AR, the authors either engineered the receptor with a T4 lysozyme in the third intracellular loop or coupled it with an antibody. Although the overall structures are similar to rhodopsin, there are apparent differences in the second extracellular loop and the E/DRY interactions between the cytoplasmic ends of transmembrane TM3 and TM6. The highly conserved E/DRY motif is known to play a critical role in regulating GPCR conformations (28). Also significant helical shifts are observed as compared to rhodopsin. In the high-resolution structure (24, 25), β2AR binds to a diffusible ligand — carazolol. Interestingly, functional characterization of this engineered β2AR shows both the wild-type-like binding to an (carazolol), and an increased affinity for agonists, which is a profile similar to that of constitutively active mutants. The snapshot of these receptors is only the first step; future studies incorporating agonist-bound and G protein- or arrestin-

2 associated GPCR structures will provide further insights into the mechanisms of GPCR signaling.

1.1.3 G Protein-Coupled Receptor Function

GPCRs transmit signals from extracellular stimuli to intracellular G proteins through

conformational changes, which demonstrate the allosteric nature and conformational

plasticity of GPCRs (29, 30). The G proteins, composed of alpha (Gα), beta (Gβ), and

gamma (Gγ) subunits, exist in a heterotrimeric complex when inactive. Signaling is

initiated by ligand-GPCR (complex) interaction, which leads to conformational changes

of the receptor and these structural changes promote the exchange of guanine-nucleotide

on the Gα subunit. The ligand bound receptor promotes Gα to release the constitutively-

bound guanosine diphosphate (GDP) and in turn bind to guanosine triphosphate (GTP).

The GTP-bound Gα subunit then dissociates from the heterotrimeric G protein-complex,

while Gβ and Gγ remain associated. Both the GTP-bound Gα subunit and Gβγ dimer can

modulate a number of downstream effectors (Figure 1.1), including phospholipases (e.g.,

phospholipase C (PLC)) and nucleotide cyclases (e.g., adenylate and guanylate cyclases).

These effectors will further produce second messengers (e.g., Ca2+ and 3’-, 5’-cyclic adenosine monophosphate (cAMP)) and eventually cause a series of biochemical and

electrical events in the living cells (31-34).

GPCR signaling is precisely regulated by a series of cellular processes, including phosphorylation, desensitization, internalization, downregulation, recycling and degradation (35, 36). It is well-known that a long period of exposure to the ligands causes

3

GPCR desensitization. Usually phosphorylation is thought to be the first step of desensitization, followed by internalization and downregulation. Several protein kinases have been shown to phosphorylate the intracellular domains of GPCRs following agonist exposure, including the second messenger-dependent kinases such as protein kinase A

(PKA) and protein kinase C (PKC), as well as specific G protein-coupled receptor kinases (GRKs) (37-41). In the classic model of β2AR signaling (Figure 1.2) (42), the binding of agonist to receptor results in the rapid phosphorylation of β2AR, which promotes the association of arrestin. The binding of arrestin to the receptor not only attenuates β2AR signaling due to the G protein uncoupling, but also induces internalization by clathrin-mediated endocytosis. The internalized GPCRs can traffic within the cell to the lysosome for degradation, or recycle to the plasma membrane in a process known as resensitization. An alternative pathway of GPCR internalization was also suggested to occur via caveolae (43-45).

Numerous theories have been proposed to explain the mechanism of GPCR activation and agonist effects (46-48). In the basic two-state receptor theory, only agonist and

receptor were thought to play an active role for signaling; whereas G proteins are only

considered to couple to GPCR after agonist-GPCR binding. However, in real biological

systems, GPCR signaling is much more complex. Without the consideration of GPCR-

accessory proteins, it is difficult to explain the multiple binding and signaling properties

of GPCRs. Accumulating evidence suggests that at least some GPCRs are actually pre-

coupled to G proteins, which allosterically affects ligand binding (29, 49, 50). Also there

is evidence showing some GPCRs can signal without the participation of G proteins (51).

4

In addition, there are other challenges in understanding GPCR structure and function, for example, the existence and significance of homo- and/or heterodimers (21, 49, 52, 53), conformational plasticity (30, 54, 55) and (54, 56). Even with some spectacular advances in recent years, GPCR signaling is still far from being completely understood.

Figure 1.1 GPCR signaling. GPCR activated by an agonist transduces signal to the G protein, which promotes the exchange of GDP to GTP in Gα subunit and dissociation of Gα subunit from Gβγ. Both the GTP-bound Gα subunit and the Gβγ dimer can activate various effectors. Typically Gαs stimulates adenylyl cyclase (AC) and increases cAMP level, whereas Gαi inhibits AC and decreases cAMP level. Gαq family activates PLC, which cleaves phosphatidylinositol biphosphate (PIP2) into diacylglycerol (DAG) and inositol triphosphate (IP3). The Gβγ dimer can activate many signaling molecules, such as phospholipases, ion channels and lipid kinases. [Modified from Figure 1 in Ref (57).]

5

Figure 1.2 GPCR signaling regulation based on β2AR model (42). Following receptor activation, GPCRs are phosphorylated by kinases, which lead to rapid desensitization due to G protein uncoupling. Then arrestin binds to the phosphorylated third intracellular loops and carboxy-terminal of the activated GPCRs. Desensitized receptors are internalized through clathrin-coated vesicles. These clathrin-coated vesicles can fuse with early endosomes where the receptors are dephosphorylated and recycled back to the plasma membrane or may be guided to lysosomes for degradation. [Modified from Figure 1.1 in Ref(58).]

1.2 Opioid Receptors

1.2.1 Opioid Receptor Overview

Opioid receptors belong to the rhodopsin-like Class A GPCRs (59, 60). Based on the

early studies of ligand binding, opioid receptors were classified into three subtypes: µ-

opioid receptor (MOR), δ-opioid receptor (DOR) and κ-opioid receptor (KOR) (61, 62).

MOR and KOR were named after the prototypic drugs, i.e. morphine and

ketocyclazocine respectively (63, 64). DOR (δ, for deferens) was named because it was

discovered in guinea-pig and mouse vas deferens (64). Later, an orphan receptor

6

(opioid receptor-like, ORL1) was placed within the opioid family because of a high degree of structural homology ( > 60%) to other opioid receptors (65). However, the , , which binds to MOR, DOR and KOR receptors with differing affinities, does not have significant affinity for ORL1. From a pharmacological perspective, ORL1 is structurally related to the endogenous opioid systems, yet is pharmacologically distinct. All the discussions in this dissertation will focus on the main subtypes of opioid receptors, i.e., MOR, DOR and KOR. The nomenclature MOR, DOR and KOR are recommended by the International Narcotics Research Conference (INRC).

There are several revisions of the terminology for opioid receptors in history. Originally opioid receptors were named by using the Greek letters µ, δ and κ, then the MOR, DOR and KOR recommended by INRC, and more recently using MOP, DOP and KOP by

IUPHAR. Because of its wide acceptance in opioid field, the INRC version is used throughout this dissertation.

The human opioid receptor were discovered and cloned in the early 1990s (66-69).

The opioid genes are located on three different and all have multiple introns. Generally, opioid receptors are about 60% homologous, with the most conserved amino acids found in the transmembrane domains and intracellular loops (70).

Subdivisions of the individual receptor subtypes ( µ1, µ2, δ1, δ2, κ1, κ2 and κ3 ) were thought to exist because of the different pharmacological properties exhibited by ligands at the same receptor subtype (59). However, cloning efforts for these putative subdivisions have not confirmed any new genes. Several mechanisms could explain the subdivision-like pharmacological profiles: receptor splice variants, single nucleotide

7 polymorphisms (SNP), heterodimerization and interactions with different accessory proteins (60, 71, 72).

Studies examing opioid receptor mRNA expressions have determined that all three classic opioid receptors are expressed throughout the central nervous system, peripheral sensory and autonomic nervous systems. Traditionally, opioids have been though to exert antinociceptive effects via the central nervous system (CNS). Now it is accepted that the opioid receptors’ action can also be peripheral and local. The presence of opioid receptors in dorsal horn and peripheral nerve terminals provides an opportunity to design peripheral without severe CNS side effects. In addition to the analgesia, opioid receptors can regulate many other physiological functions. MOR can regulate respiratory and cardiovascular functions, mood, thermoregulation, secretion and immune functions (73). KOR causes with limited liability, while , respiratory and physical dependence mainly involve MOR and

DOR. Additionally MOR and KOR are found in the wall of the (74-

77). KOR was also suggested to be expressed on CD4+ lymphocytes and monocular phagocytes, the major immune cell types infected by HIV, because of the suppression of

HIV-1 expression by KOR selective agonists (78). Data from opioid receptor knockout animals have confirmed many of these activities of opioid receptors (70).

1.2.2 Opioid Receptor Function

Opioid receptors are prototypical Gαi/o coupled receptors and thus pertussis toxin (PTX)

sensitive. Opioid receptors are known to promiscuously couple to all Gαi/Gαo subtypes

8 and the PTX-insensitive Gαz protein. Like many other Gαi/Gαo coupled receptors, the

opioid receptors can inhibit adenylyl cyclases and Ca2+ channels, and stimulate K+ channels. Lately the opioid receptors have been shown to regulate the mitogen-activated protein kinase (MAPK) cascade. The inhibition of adenylyl cyclases leads to the decrease of cAMP production, which reduces the level of release by attenuating the activity of cAMP-dependent PKA (79, 80). There is accumulating evidence that the release of glutamate, γ-aminobutyric acid (GABA) and throughout the CNS can be inhibited by opioid receptors (81). For example, KOR agonists are dysphoric because of the direct inhibition of release from nerve terminals in the nucleus accumbens (NA). Although the predominant action of opioids in the nervous system is inhibitory, in several brain regions important for either supraspinal analgesia (e.g. periaqueductal grey (PAG)) or euphoria (e.g. ventral tegmental area (VTA)) are excitatory. It is now accepted that opioid-induced excitations are due to a disinhibition mechanism (Figure 1.3), which is the inhibition of inhibitory .

Other than the Gα subunit, opioid receptors also can signal through Gβγ subunits (82) to cause the activation of K+ conductance, inhibition of Ca2+ channels and modulation of the

N-methyl-D-aspartate (NMDA) receptors. It is suggested that opioid receptors can

regulate the PLCβ signaling pathway and Ca2+ mobilization via Gβγ, without involving

Gαq (83). However, other mechanisms may still play a role; for example, both DOR and

KOR were shown to stimulate PLCβ by coupling to Gα16 protein (84). The activation of

MAPKs can also occur through the Gβγ subunits (85). Gβγ subunits primarily regulate the MAPKs through their effector-mediated pathways. It is very likely that opioid

9 receptors are able to mediate cellular responses by several mechanisms and the relative importance of each mechanism may depend on the cellular milieu.

Figure 1.3 Opioid receptors regulate the dopaminergic activity in VTA through the disinhibition mechanism. Stimulation of opioid receptors reduces γ-aminobutyric acid (GABA) transmission, which increases the firing rate of dopaminergic via disinhibition. [Modified from Figure 2 in Ref (86).]

1.2.3 Opioid Receptor Signaling Regulation

10

Opioid receptors are subject to agonist-activated phosphorylation by GRKs and activation of the arrestin-mediated endocytic pathway, as discussed previously (Figure 1.2, Section

1.1.3). The mechanism of desensitization is that agonist-activated receptors are first phosphorylated by GRKs, which then facilitates arrestin binding and prevents the receptor from coupling to G proteins (42, 87, 88). Arrestin-bound receptors are internalized and sorted in early endosomes and then may traffic to either lysosomes for downregulation or into endosomes for dephosphorylation and recycling, also known as resensitization. The major phosphorylation sites are in the carboxyl tail and intracellular loops of the opioid receptors. Besides GRKs, other protein kinases might be involved in the phosphorylation of the opioid receptors, such as ERK1/ERK2. Receptor internalization also appears to be agonist and receptor type dependent. An interesting example is the case of morphine and DAMGO; even though both are high affinity MOR agonists, DAMGO, but not morphine, can induce MOR phosphorylation (89). Recent studies have shown that opioid receptors, like some other G protein-coupled receptors, can form functional homo- or heterodimers (90, 91). It is still a challenge to characterize different forms of opioid receptor complexes, including the homo- and heterodimers.

1.3 Opioids

1.3.1 Endogenous Opioids

The term opioid encompasses all morphine-like agonists as well as endogenous opioid

peptides. Opioid receptors are activated physiologically by their endogenous opioid

peptides (Table 1.1). Endogenous opioids play a critical role in modulating pain,

perception, mood, reward and some autonomic nervous system functions. Two closely

11 related endogenous opioids (Met- and Leu-enkephalin) were first identified by

Hughes et al. in 1975 [NOTE: The name of enkephalin is derived from the Greek en kephalos, meaning in the head]. The precursor protein, , gives rise to the endogenous opioids after proteolytic cleavage (59). Met-enkephalin and Leu-enkephalin have differential affinities for opioid receptors, from high affinity to low affinity in order of DOR>MOR>>KOR. Other precursor proteins and products are subsequently found and characterized (Table 1.1). generates β-endorphin, which is an agonist at both MOR and DOR, with little affinity for KOR. can produce multiple dynorphin peptides (dynorphinA and B, α-neo-endorphin, and β-neo-endorphin), which have highest affinity for KOR but also bind to MOR and DOR. -1 and -2, whose precursor is still unknown, show high selectivity for MOR (92). All these peptides are full agonists for their cognate receptors, but none of them is exclusively selective (65) and none of these peptides have apparent affinity for ORL1. Similarly, the recently discovered ORL1 peptide agonist, /orphanin-FQ, has no apparent affinity for MOR, DOR and KOR.

Table 1.1 Endogenous opioid peptides*

Opioid peptide Precursor sequence Receptor type product MOR and Proenkephalin [Met]-enkephalin YGGFM DOR (93) MOR and [Leu]-enkephalin YGGFL DOR (93)

YGGFMTSEKSQTPLVTLF MOR and Proopiomelanocortin β-endorphin KNAIIKNAYKKGE DOR (94, 95)

Prodynorphin dynorphin A YGGFLRRIRPKLKWDNQ KOR(96)

12

YGGFLRRQFKVVT KOR (71) α-neo-endorphine YGGFLRKYPK KOR (71) β-neo-endorphine YGGFLRKYP KOR (71) Pronociceptin and nociceptin and FGGFTGARKSARKLANQ ORL1(95, 97) orphanin-FQ orphanin-FQ

endomorphin-1 YPWF-NH2 MOR (92) Unknown endomorphin-2 YPFF-NH2 MOR (92)

Prodermorphin and Y(D)AFGYPS-NH2 MOR (71)

prodeltorphin Y(D)MFHLMD-NH2 DOR (71) *[adapted from Table 1 in Ref (71) and Figure 3 in Ref (59)]

1.3.2 Exogenous Opioids

Opium, the extract of the poppy plant, has been used medicinally for hundreds of years.

In the early 1800s, the active pharmacological ingredient of the poppy plant was

identified and named morphine after Morpheus, the Greek god of dreams (98, 99).

Morphine and synthetic morphine-like , MOR agonists, can effectively reduce

severe pain. However, MOR agonists have severe side effects including respiratory

depression, tolerance and dependence. The major goal of opioid research is to discover

new drugs without all or some of those side effects, thereby avoiding opioid . Two approaches have been applied to design and improve opioid ligands: 1) modifying morphine and 2) modifying endogenous peptides. The latter approach would also target nonpeptide peptidomimetics for better bioactivity and selectivity. After over

50 years of synthetic efforts, thousands of compounds based on natural structures have been developed. Sadly, it has now become clear that analgesic potency and dependence liability are inseparable for MOR agonists. The progressive simplification of morphine structure is shown in Figure 1.4. From the , the

13 , the piperidines, to the phenylpropylamines, numerous compounds with unique properties are still widely used in the clinic (Table 1.2).

N

O Morphine derivatives e.g.

N N

O derivatives derivatives e.g. e.g.

N N N

Phenylpropylamine derivatives Piperidine derivatives derivatives e.g. e.g. , e.g. , ketocyclazocine,

Figure 1.4 Chemical backbones of morphine-derived molecules. [Adapted from Figure 4

in Ref (71) and Figure 4 in Ref (59).]

Table 1.2 Select list of endogenous opioid peptides and morphine derivatives.

Receptor Name and structure Precursor Function type

N H Selective - MOR agonist HO O OH Morphine

14

N H Selective Morphine MOR agonist AcO O OAc

N H Selective Morphine MOR agonist MeO O OH Ph

N Selective Piperidine MOR N agonist O Et Fentanyl S

N Selective Piperidine MOR agonist N OMe O Et

N

O Selective Phenylpropylamine MOR Et agonist

Methadone OH

H O O Selective N N OH Enkephalin MOR H2N N N agonist O H O H DMAGO

N OH MOR, DOR, Non-selective Morphine KOR antagonist

HO O O Naloxone

15

N OH Non-selective Morphine MOR antagonist

HO O O Tyr-D-Ala-Gly-Phe-D-Leu-OH Selective Enkephalin DOR DADLE agonist Tyr-D-Ser-Gly-Phe-D-Leu-Thr-OH Selective Enkephalin DOR DSLET agonist SS Selective Tyr-D-Pen-Gly-Phe-D-Pen-OH Enkephalin DOR agonist DPDPE

Tyr-D-Ala-Gly-Phe-D-Leu-OH Selective Enkephalin DOR DADLE agonist SS Selective Tyr-D-Pen-Gly-Phe-D-Pen-OH Enkephalin DOR agonist DPDPE

H N

N Selective Morphine DOR agonist OH TAN-67

N OH Selective Morphine DOR antagonist

HO O N H (NTI)

N OH Selective Morphine DOR antagonist

HO O Bn N-Benzylnaltrindole (BNTI)

16

N

O Selective Cl N Arylacetamide KOR Cl agonist U50488

N

O O Selective Cl N Arylacetamide KOR Cl agonist U69593

OH O N Selective Benzomorphan KOR HO agonist Et Bremazocine

N OH Selective Naltrindole KOR agonist NH O HO N N NH2 6'-GNTI H H

Selective - KOR agonist

N OH Selective H Naltrindole KOR NNH antagonist

NH HO O N 2 5'-GNTI H

N N OH HO Selective Naltrindole KOR antagonist N O H O HO OH norBNI

17

1.3.2.1 MOR Ligands

Thousands of morphine analogs have been synthesized, some representative compounds are shown in Table 1.2. Here is a brief introduction of some typical morphine-like compounds. is a nonselective agonist for all opioid receptors and is a potent analgesic with severe side effects. Etorphine is only used for immobilizing large animals

(100). Buprenorphine, a partial MOR agonist and a KOR antagonist, is a powerful analgesic for the treatment of moderate and severe pain (101). is an antagonist at the MOR and an agonist at KOR (102). Nalorphine is a strong analgesic with limited respiratory depression; however, it has dysphoric and effects. Fentanyl, and analogues such as sufentanil and , are among the most potent MOR agonists and widely used for surgical anaesthesia (60, 102). Another very important morphine derivative is methadone, which is widely used to treat opioids addiction and withdrawal (103). Naloxone and naltrexone are non-selective antagonists for all opioid receptors (104).

1.3.2.2 DOR Ligands

Most DOR ligands are tools for opioid research, having no medicinal use. Two selective

DOR agonists, [D-Ala2, D-Leu5]-enkephalin (DADLE) and [D-Ser2, Leu5]-enkephalin

(DSLET), are peptides. Conformationally constrained cyclic peptides such as [D-Pen2,

D-Pen5]-enkephalin (DPDPE) is a potent DOR agonist (60). The nonpeptide selective

DOR antagonist, naltrindole (NTI), is an analogue of naltrexone. NTI leads to the discovery of the benzyl derivative (N-benzylnaltrindole (BNTI). These compounds are currently used as selective DOR antagonists. Other nonpeptide DOR agonists include

18 morphinan derivatives TAN-67. The derivative SNC-80 represents a new series of agonists (105). DOR agonists show both antinociceptive and -like activities in animal behavior models (106, 107), although the proconvulsant effects limit their therapeutic use.

1.3.2.3 KOR Ligands

Similar to DOR ligands, KOR ligands are also mainly used for research; however, some

novel KOR ligands show potential for medicinal applications (108). Chemically, KOR

agonists can be categorized in five classes: the endogenous peptides (dynorphin), the benzomorphans, the arylacetamides (prototype U50488 and U69593), the derivatives and lastly salvinorin A. The benzomorphan compounds, which include ethylketocyclazocine and bremazocine, show marginal KOR selectivity. They have been rejected from clinical development due to psychotomimetic and dysphoric effects, despite low dependence liability (109). Because it was thought that the severe side effects of opioid drugs were due to the lack of selectivity, there was some optimism in truly selective KOR agonists. However, this hope failed after testing a novel class of selective

KOR agonists — the arylacetamide derivatives including U50488, U69593 and

(110). Another similar compound, TRK-820, is a morphinan derivative with stronger analgesic effect. Recently, salvinorin A, the main active ingredient of Salvia divinorum

(S. divinorum), was identified as a novel KOR agonist (1). Salvinorin A has no structural resemblance to any known opioid ligands (Table 1.2). In recent years, a number of KOR antagonists have been designed based on the DOR antagonist NTI (111). The bivalent

NTI ligand, nor- (nor-BNI), has been identified as a highly active and

19 selective KOR antagonist. Additional derivatives, such as 5’-GNTI and 6’-GNTI, have been designed based on nor-BNI. Because adverse CNS effects are associated with many

KOR agonists, recent efforts have focused on peripherally acting agents (112-114).

1.3.3 Opioids’ Medical Use and Abuse

Opioids are the most effective analgesics (103, 115). Conventionally, opioids, such as the

MOR full agonists (morphine, , oxycodone, , etc.), have

been used to treat both moderate and severe pain. Although these drugs have their

limitations, no other group of drugs has been found to provide better pain relief. To meet

various clinical needs, both immediate-release (short-acting) and controlled-release

(long-acting) opioids are produced. For postsurgical pain, short-acting opioids with

duration of 2 to 4 hours are preferred. For example, the ultra-short-acting analogue of

fentanyl, remifentanyl, is rapidly metabolized by blood and tissue esterases with a half-

life shorter than 10 min (116). Conversely, cancer pain, chronic and nonmalignant pain

require long-acting opioids with a duration of 12-24 hours (117).

Opioid abuse has accompanied their medical uses since the beginning. Many opioids are

on the controlled substance list of the Drug Enforcement Administration (DEA). For

example, heroin is on schedule I, morphine and fentanyl are on schedule II. It is also

well-known that opioid drugs cause tolerance and physical dependence. Of these three

opioid receptors, MOR is responsible for the reinforcing effect of opioids. MOR agonists

induce euphoria by indirectly enhancing dopamine in the NA. The mechanism is by

inhibiting GABA release within the VTA (Figure 1.3), thus disinhibit the dopaminergic

20 neurons that project to NA. Similar to , and , morphine can cause long-term potentiation (LTP) of transmission onto VTA dopaminergic neurons; however, the mechanisms are currently unclear (118). LTP is thought to underlie learning and memory and LTP induction by opioid drugs may explain opioid addiction. Other mechanisms could also play a role here, such as changes at the level of the opioid receptors (phosphorylation, desensitization, interactions with other receptors and G protein-uncoupling), the alteration of gene expression and changes in neuronal circuits (70). There is still controversy about the role of desensitization in morphine tolerance (72, 81, 87).

Another phenomenon associated with opioid dependence is the severe withdrawal effect after chronic opioid use. Biochemically, there is a down-regulation of MOR and an up- regulation of cAMP-PKA pathway. It is believed that withdrawal is caused by the enhancement of cAMP production and consequently increased neurotransmitter release

(81). MOR agonists, such as methadone, are the main pharmacological treatments for . Methadone produces similar analgesia and side effects as morphine, but also has a long duration of action and slow elimination; meanwhile, it presents slow and relatively mild withdrawal effects (119, 120). The related compound, L-alpha-acetyl- methadol (LAAM), having a longer action than methadone, was approved by the FDA for opioid abuse treatments in 2002.

1.4 The History of Salvia Divinorum and Salvinorin A

Research

21

S. divinorum, or Diviner’s sage, is a powerful psychoactive herb belonging to the

Lamiaceae (mint) family. S. divinorum was first recorded in print in 1939 (121), and introduced to western countries by Wasson and Hofmann in 1962 (122, 123). Before then it had been used as an ritual entheogen by Mazatec shamans for centuries (124). The origin of the S. divinorum is not very clear. A commonly accepted opinion is that S. divinorum is a cultigen created by Mazatec people, not occurring in the wild state. S. divinorum was used traditionally to produce a hallucinogenic experience essential for divination and healing. Over the past two decades, S. divinorum has been increasingly used for recreational purpose due to the easy access through the . Some Internet- based companies sell live Salvia plants, dried leaves, as well as extracts. Even though S. divinorum has some hallucinogenic effects like lysergic acid diethylamide (LSD), it also differs from the common recreational drugs, such as LSD, marijuana or mushroom

(containing ) (125). S. divinorum is legal in most countries and in the majority of states within the USA (www.erowid.org). In 2003 DEA placed S. divinorum on the list of drugs of concern. Many scientists who see the importance of salvinorin A in research support regulating Salvia as an entheogen rather than as a LSD-like hallucinogen

(www.sagewisdom.org). Generally entheogen is a psychedelic substance used in a religious context to bring on a spiritual experience (126-128). It is of great interest to keep Salvia and its major active component — salvinorin A available in opioid research, which already showed great potential for medicinal applications and as a research tool to explore the function of opioid receptors.

22

The neoclerodane diterpene salvinorin A is the primary psychoactive component of S. divinorum. Salvinorin A was first isolated from S. divinorum leaves and chemically characterized in 1982 (129). Initial attempts in the 1990s to identify the molecular target of salvinorin A failed. Salvinorin A was submitted to the National Institutes of Mental

Health sponsored Psychoactive Drug Screening Program (PDSP) in 2001 and screened against more than 50 targets including GPCRs, ion channels and transporters.

Surprisingly, salvinorin A was found to only bind to KOR and not even its close family members MOR and DOR (1). Interestingly, it has no actions at the 5-HT2A receptor, the principal molecular target for classic such as and LSD. Neither does salvinorin A bind to NMDA and PCP sites. The most convincing proof that the

KOR is the site of action is that the effects of salvinorin A are blocked by selective KOR antagonists in mice (130). The scientific research of salvinorin A accelerated after the discovery of KOR as the molecular target. Salvinorin A has been confirmed as a highly selective and potent KOR agonist with convincing proof both in vitro and in vivo (1,

125), some of which is shown in Table 1.3.

Salvinorin A is unique in its structure (Table 1.2) and function (Table 1.3). Unlike other known opioid-receptor ligands, salvinorin A is not an alkaloid — it does not contain a basic atom (1, 131). Therefore, it represents the only known psychoactive diterpene and non-nitrogenous hallucinogen. Also, salvinorin A is the most potent naturally-occurring psychoactive compound with doses as low as 200 µg. The effective dose of salvinorin A in humans approaches that of the synthetic hallucinogen

LSD and 4-bromo-2,5-dimethoxy-phenylisopropylamine (DOB). In addition, the

23 antinociceptive effect of salvinorin A was examined in both regular and KOR knockout mice by tail-flick latencies or hot-plate assays. The antinociceptive effect diminishes within 30 minutes after intraperitoneal injections (IP) of salvinorin A (1-4 mg/kg) (132).

Several publications investigated the analgesic effects of salvinorin A (133-135). As a

KOR agonist, salvinorin A could also have dysphoric and depressive-like effects (136,

137). -like effect of salvinorin A was observed in a forced swimming test

(137), which is a well-accepted rodent behavior model for the evaluation of antidepressant drug effects. IP of salvinorin A in rats increases immobility in the forced swim test. On the other hand, KOR antagonists could have antidepressant-like effects in vivo (138, 139). Additional studies in mice show that salvinorin A indeed decreases the dopamine level in the caudate , but not in NA (130). These observations confirm that salvinorin A acts on KOR and provide a new treatment of depression by modulating KOR signaling pathways. Other therapeutic potentials of salvinorin A and related compounds are treating cocaine addiction and psychiatric disorders associated with , such as and Alzheimer’s disease

(108, 140-144). The toxicological study of salvinorin A showed no signs of organ damage in rodents after a two-week treatment (145).

Table 1.3 Pharmacological effects of salvinorin A in select animal studies*

Drugs Animal With Effects Dose* Methods Antagonist* Ref Model Similar Effect Decreases basal U69593, 1.0 mg/kg Blocked by dopamine mice U50488 and 3.2 nor-BNI, (130) leveal in and HPLC (C57BL/6J) and mg/kg, IP 10 mg/kg caudate R84760

24

putamen, but not in nucleus accumbens U69593, Induces place Conditioned 1.0 mg/kg Blocked by U50488 aversion and place mice and 3.2 nor-BNI, and (130) decreases preference (C57BL/6J) mg/kg, IP 10mg/kg TRK- locomotion chamber 820 Monkeys trained to Fully blocked 0.001- discriminate Produces drug by Rhesus 0.032 U69593 from (0.32mg/kg), monkey U69593 (146) discrimination mg/kg, SC vehicle by and partially by food GNTI(1mg/kg) reindorcement Pharmcokinetic Plasma 0.032 Rhesus elimination t1/2= concentration - monkey - (147) mg/kg, IV 56.6 ± 24.8 min by HPLC Thermal (tail- flick) and Dose dependent chemo(acetic and time 0.5-4 acid Nor-BNI mice - (132) dependent mg/kg, IP abdominal 10 mg/kg antinociception constriction)- nociceptive assays Induces Inverted sedation and 0.5-2 screen Nor-BNI, mice U69593 (148) motor mg/kg, IP performance 30 mg/kg incoordination assay 1 pM- Electrical field Inhibit enteric 1 µM, stimulation Nor-BNI, 30 nM and Guinea-pig - (135) incubation induced transmission Naloxone 1 µM with tissue contraction Nor-BNI, Dose dependent 14-23 Tail-flick 6.8 nmol; but not NTI mice - (149) antinociception nmol, IT latency and β- funaltrexamine rats trained to discriminate Produces drug 1-3 U69593 from Nor-BNI, 4.5 rats U69593 (150) discrimination mg/kg, IP vehicle by nM, ICV food reindorcement Mixing Swimming test Nor-BNI, stimulating and 0.1-10 and 10 mg/kg; zebrafish - (151) depressive µg/kg, IM conditioned , effects place 1 mg/kg

25

preference

*IP, intraperitoneal injection; SC, ; IV, intravenous injection; IT, intrathecal injection; ICV, intracerebroventrular injection; IM, .

Dr. Bryan Roth’s lab is one of the major forces pushing the study of salvinorin A forward since its discovery of KOR as the molecular target in 2002 (1). Through the efforts of several research groups, we now have a better understanding of the actions of salvinorin

A on KOR. Salvinorin A has also been proven to be a valuable research tool to study opioid receptors. The long term goal of salvinorin A research is to discover new therapeutic agents without the and hallucinogenic effects. In the following chapters, a combination of molecular pharmacology, biochemistry, chemistry, computer modeling, mass spectrometry and animal behavior methods have been applied to the study of salvinorin A. Specifically, I have proposed three specific aims for my dissertation research: specific aim 1 was designed to identify the binding site of salvinorin A in KOR; specific aim 2 was designed to investigate KOR’s conformational change during the activation process; specific aim 3 was to further verify salivinorin A -

KOR interactions by designing a series covalently-bound KOR ligands based on salvinorin A’s structure. During the past four and half years, eight research papers (2-7,

152, 153) and one review paper (108) related to my work have been published in peer- reviewed journals. Step by step, the molecular mechanisms by which salvinorin A binds to and activates KOR were revealed. Correspondingly, I will report our findings in three parts (Chapter 3, Chapter 4 and Chapter 5) in this dissertation. The future direction of salvinorin A/KOR research will be discussed in chapter 6.

26

Chapter 2: Materials and Methods

2.1 Materials

2.1.1 Chemicals 2.1.1.1 Commercial chemicals

Standard reagents were purchased from Sigma-Aldrich (St. Louis, MO).

[3H]diprenorphine (54.9 Ci/mmole), [3H]U69593 (41.7 Ci/mmole) [3H]DADLE (44.3

Ci/mmole), and [3H]DAMGO (56.8 Ci/mmole) were purchased from PerkinElmer-

LifeScience, Inc. MTSEA was obtained from Anatrace, Inc. In binding site study (2), two

sources of salvinorin A were used: Biosearch and the S. divinorum Research and

Information Center, Malibu, CA. For the conformational study (7), salvinorin A was

kindly provided by Dr. Thomas Prisinzano (University of Iowa). Naloxone and U69593

were purchased from Sigma-Aldrich. Dynorphin A (1-13) was purchased from either

Sigma-Aldrich or Bachem Bioscience, Inc.

2.1.1.2 Synthesized salvinorin A derivatives

This part of the work was done by collaborating with two labs, Dr.

Jordan K. Zjawiony (University of Mississipi) and Dr. Mark A. Rizzacasa (The

University of Melbourne, Australia). There are 55 salvinorin A derivatives have been

tested in our lab (Table 3.6 in Chapter 3). The chemical syntheses and characterizations

for half of these compounds have been published in recent papers (2-6), are not repeated

here.

27

2.1.2 cDNA Constructs 2.1.2.1 Sub-cloning of hKOR into pUniversal-signal

Dr. Wesely Kroeze in Dr. Bryan Roth’s lab constructed a pIRES-neo (Clontech, Palo

Alto, CA) based, membrane protein expression vector—pUniversal-signal(154). It

contains a Kozak sequence (CACCATG) to enhance (155); an N-terminal hemagglutinin signal sequence to enhance the translocation of receptor to the plasma membrane; a FLAG epitope tag. FLAG-tagged human KOR was subcloned into the vector pUniversal-Signal. The presence of the correct KOR sequence was verified by automated dsDNA sequencing (Genomics Core facility, Case Western Reserve

University) before use.

2.1.2.2 Generation of KOR mutants by site-directed mutagenesis

Regular Mutants — Mutations were introduced using the Quickchange mutagenesis kit from Stratagene according to the manufacturer's recommendations and verified by automated dsDNA sequencing. Most mutants are listed in Table 3.1 (Chapter 3).

Mutants for SCAM — Site-directed mutagenesis was based on the C3157.38S background

of the human KOR in the vector pcDNA3.1 following the procedure of the Quickchange

mutagenesis kit. About 40 mutants (TM7, TM6 and EL2) were made for the SCAM study

(Figure 4.2, Chapter 4).

2.1.3 Cells 2.1.3.1 Transient transfection and cell membrane preparation

28

Cells (HEK 293T, Gα16 and Gαi2 Cells) were grown in 10 cm culture dishes in medium

with 10% fetal calf serum in a humidified atmosphere consisting of 5% CO2 and 95% air

at 37 °C. Cells were transfected with either the wild-type, C3157.38S hKOR single mutant,

or a double mutant hKOR DNA incorporating the C3157.38S mutation (12 µg/10 cm dish)

using EasyTransgater (America Pharma Source). After a total of 48 hours of transfection, cells were harvested for experiments by detaching with Versene solution (Invitrogen).

HEK 293T cells are a highly transfectable derivative of the HEK (human embryonic kidney) cell line, which has the temperature sensitive gene for SV40 T-antigen.

The transient transfected or stable cells were scraped and centrifuged at 1000g for 5 min, then cell membranes were lysed in standard binding buffer (50 mM Tris, 0.1 mM EDTA,

10 mM MgCl2, pH 7.4) and cell lysates were centrifuged at 14000g for 20 min to get membrane pellets. The membrane pellets were washed in the standard binding buffer and kept at -80 °C until use in binding assays.

2.1.3.2 Stable expression of G proteins

Stable cell lines expressing the human Gα16 and Gαi2 were obtained by transfecting the G

protein expression vectors (pcDNA3.1, UMR cDNA Resource Center) into HEK 293

cells and selecting in 900 µg/mL G418. Those cell lines were maintained and transfected

in 450 µg/mL G418. The expression of Gα16 and Gαi2 were characterized with anti-

human Gα16 and Gαi2 polyclonal antibodies (Cell Sciences, Inc.).

2.2 Methods

29

2.2.1 Radioligand Binding Assays 2.2.1.1 Competition binding assay

The binding assay for KOR uses either [3H]diprenorphine or [3H]U69593 (PerkinElmer-

LifeScienceInc) as radioligand. Ki determinations were performed by using 10

concentrations of unlabeled ligand spanning an appropriate dose range (~105 to 10-3 nM).

All assays were conducted in triplicate or duplicate using polypropylene standard (8 x 12 format) 96 deep well (1 mL/well) plates as binding reaction containers. Reactions were in

250 µL and contained 175 µL of standard binding buffer, 25 µL of test drug or buffer (for total binding), 25 µL of [3H]diprenorphine (0.15 ~ 0.25 nM final concentration), and 25

µL of membrane receptors. Nonspecific binding was defined by Naloxone at a final

concentration of 10 µM (SigmaAldrich). The 96-well plates were incubated in the dark at

room temperature for 90 min. Filters were presoaked in 0.3% PEI in 50 mM Tris buffer

(4 °C, pH 7.4). The binding reaction was terminated by rapid filtration under vacuum by a Brandel Harvester (Brandel). Each well was washed three times with 0.5 mL of cold 50 mM Tris buffer (pH 7.4). Filters were dried and placed into 6.5 ml scintillation vials

(Laboratory Products Sales Inc). To each vial, 4 ml Ecoscint A biodegradable scintillation solution (National Diagnostics) was added. After being capped, those scintillation vials were counted in a liquid scintillation counter (PerkinElmer-

LifeScienceInc). Each vial was counted for about 2 min. The raw data were analyzed by

Prism (GraphPad Software, Inc.) to generate Ki values reported as the mean ± standard

error of the mean (SEM).

2.2.1.2 Saturation binding assay

30

Saturation binding was used to determine Kd and Bmax Values for WT KOR and mutants.

Membranes were prepared from transfected HEK 293T cells. Saturation binding of

[3H]diprenorphine was performed at eight concentrations of [3H]diprenorphine ranging

from 0.03 nM to 3 nM. Binding was carried out in standard binding buffer at room

temperature for one hour in triplicate in a volume of 250 µL with about 30 µg of

membrane protein. Naloxone (10 µM) was used to define nonspecific binding. Protein

concentrations of membranes were determined by the Bradford protein assay method with BSA as the standard. Binding data were analyzed with Prism (GraphPad Software,

Inc.).

2.2.2 SCAM

2.2.2.1 MTSEA reaction

Transfected cells were incubated with 2ml/plate Versene solution (Invitrogen) for 2 min,

detached and pelleted at 1000g at 4 °C. After being washed with cold Kreb buffer (130 mM NaCl, 4.8 mM KCl, 1.2 mM KH2PO4, 1.3 mM CaCl2, 1.2 mM MgSO4, 10 mM glucose, and 25 mM HEPES at pH 7.4), the pellets were centrifuged and resuspended.

The cell suspension was incubated with freshly prepared 1 mM MTSEA in 0.125 mL at room temperature for 5 min. The reaction mixtures were quenched by ice-cold 0.8% BSA in Kreb’s buffer, then pelleted and washed with regular cold buffer. After centrifugation, the pellets were resuspended and 100µL aliquots were used for [3H]diprenorphine

binding.

Rationale: The MTS reaction mechanism is a nucleophilic reaction (Fig 2.1), nucleophilic

sulfhydryl group (SH or likely S-) will replace methanethiosulfonate group and build a 31 new disulfide bond. SCAM utilizes the specific reaction between SH and methanethiosulfonate reagents (MTS) to detect the secondary structure information of the receptor protein (Fig 2.1). After the chemical modification by MTS, an observable inhibition of the receptor function is taken as evidence for this reaction, for example the interruption of normal ligand-receptor binding or change of activity. Mutation affects surface expression and surface expression level does affect the SCAM result. This is the theoretical limitation of SCAM, because the inhibition is calculated by Bmax values

(Equation 2.1). For this concern, we made Cys mutations for the whole helix 7 residues.

And our conclusion was drawn based on the SCAM pattern changes of 23 consecutive residues rather than a few specific ones.

Figure 2.1 Nucleophilic reaction between MTSEA and Cys side chain –SH (S-).

2.2.2.2 SCAM radioligand binding assay

The setup for the SCAM radioligand binding assay was similar to the above competition binding assay (section 2.2.1.1). However, each reaction contained 100 µL of standard binding buffer, 25 µL of naloxone at a final concentration of 10 µM (or buffer for total binding), 25 µL of [3H]diprenorphine (0.2~0.4 nM final concentration), and 100 µL of

membrane receptors. After counting, the raw data were analyzed by GraphPad Prism to

32 determine the inhibition number. The inhibition number was calculated using Equation

2.1(156):

Inhibition=[1-Bmax(MTSEA)/Bmax(NO MTSEA)] (2.1)

where Bmax(MTSEA) is the specific binding after the MTSEA reaction and Bmax(NO MTSEA) is the specific binding without MTSEA. Statistical comparisons were made by one-way analysis of variance (ANOVA) followed by Dunnett’s post test (using p < 0.01 as the level of significance, Figure 4.2 in Chapter 4).

2.2.2.3 Determination of second-order rate constants

The second-order rate constants of the reactions between the KOR Cys mutants and

MTSEA was determined [estimated using a pseudo-first-order equation 2.2] to gain quantitative information on Cys sensitivity, according to the published method (157,

158). Cells expressing a KOR mutant receptor were incubated with indicated concentrations of MTSEA (mostly 0.01, 0.25, 1.0 and 2.0 mM) for 5 min. The results were fit to Equation 2.2:

Y = Ne−kct + plateau (2.2)

where Y is the fraction of the initial binding, N is the extent of inhibition, k is the second-

order rate constant (M-1s-1), c is the concentration of MTSEA (M), t is the incubation time

(300 s) and plateau is the fraction of residual binding at saturating concentrations of

MTSEA. The data were analyzed by the built-in kinetic function of GraphPad Prism.

2.2.2.4 Ligand protection against the MTSEA reaction

33

This experiment is an extension of the MTSEA reaction as shown above (section 2.2.2.1).

Only one step was added: detached cells were incubated with indicated concentrations of naloxone (or U69593) for 20 min before MTSEA reaction. The calculation of protection is using Protection=[1-Bmax(Naloxone+MTSEA)/Bmax(MTSEA)] according to ref (156).

2.2.3 Functional Assays

2.2.3.1 Intracellular calcium mobilization (Ca2+ flux)

Functional assays using WT and mutant KORs were performed as previously detailed

(159). Both the stable line rKOR Cell and mutant hKOR cells (co-transfected with Ga16 into HEK-293) were plated the night before the experiment at about 50,000 cells/well

(100 µL/well of poly- treated 96-well assay plates). Cells were incubated at 37 °C overnight. The regular media was changed to serum free and indicator free media at least one hour before loading reagent buffer. 1X Reagent loading buffer was diluted from 10X component B loading buffer with a final 2.5 mM probenecid (inhibitor of the anion- exchange protein) at pH 7.4 (by 1 M NaOH). One hour before assay, the serum free buffer was exchanged to dye-containing reagent buffer (30 µL/well) for 96-well assay plate. After incubating at 37 °C for one hour, another 30 µL/well loading buffer was added to the 96-well assay plate right before starting the assay. Drug compounds plate was prepared in advance with appropriate drug concentration gradient: 0, 1, 10, 30, 100,

1000, 10000 nM. Hydrophilic drugs were dissolved directly with loading buffer.

Hydrophobic drugs were dissolved in DMSO first and diluted with loading buffer to the appropriate concentration. The 200µL 96-well tips (Molecular Devices Corp) and drug compounds plate (96-well polypropylene deep well plate 2.0mL) should be incubated to

34

37 °C before starting the FlexStationII to collect data. The samples were excited at 485 nm, and emission was detected at 525 nm. Initial volume was 60µL, and the pipettor height was set to 80 µL to transfer a volume of 30 µL at a rate of 1 (corresponds to a rate of 26 µL/second). Drug compound transfer was started after 20 seconds of background reading. Run time was set to 80 seconds to get 53 reads with a 1.52 seconds interval. The

raw data were analyzed by GraphPad Prism give EC50 and Emax values.

2.2.3.2 [35S]GTPγS binding assay

Similar as the above binding membrane preparation (section 2.1.3.1), rKOR cells were

detached in PBS, centrifuged at 1000g/10min, and washed with standard binding buffer.

Fresh membrane pellets were used for [35S]GTPγS binding assays. For each 96-well plate assay, 4-5 pellets (1 pellet from 1 10-cm dish) were needed. Ten different concentrations of test drugs in appropriate concentrations are made in the binding buffer (50 mM

Tris·HCl, 100 mM NaCl, 10 mM MgCl2, 1 mM EDTA, pH 7.4). The setup of the

[35S]GTPγS assay is in 96 well sample plates (Wallac) designed for 1450 MicroBeta

Counter (Perkin Elmer): 50 µl drug solution, 50 µl [35S]GTPγS (PerkinElmer, 1250

Ci/mmol), 50 µl membrane plus 20 µM GDP (5 min incubation), incubating the mixture

for 15 min, then 50 µl beads (FlashBlue, PerkinElmer, stock is 100 mg/ml, 110 µl stock

solution/ plate). The plates were shaken for half an hour on Titer Plate Shaker (Lab-line

Instruments, Inc.), spun down at 1000 rpm (~270g) for 2min, then counted with a

MircoBeta Counter (PerkinElmer).

2.2.4 Computer Modeling

35

This part of the work was in collaboration with Dr. Richard Westkaemper (Virginia

Commonwealth University). A short introduction is shown here, the details have been published in recent papers (2, 7, 160).

2.2.4.1 KOR Modeling for binding site (2)

Alignment of the rhodopsin and KOR sequences was performed manually by matching the highly conserved residues in transmembrane helices previously identified (161),

(162). Amino acid side-chain geometries for the KOR receptor model were established from backbone-dependent libraries of rotamer preference using the program SCWRL

(163). The PROTABLE facility within SYBYL was used to identify sites of unusual and sterically clashing side chain geometries that were interactively corrected as necessary.

Model minimizations were performed using the Tripos Force Field, Gasteiger-Huckel charges with a distance-dependent dielectric constant = 4 and non-bonded cutoff = 8 Å to a gradient of 0.05 kcal/(mol·Å). The KOR model was further modified by minimization with an ensemble of known agonists docked into the receptor (164, 165). In order to explore the inherent flexibility of the EL2 loop and the effects of the EL2 loop in limiting ligand access to Y3137.36, molecular dynamics (MD) simulations were performed. In

preparation for the MD simulations, the N-terminus of the KOR was removed so as not to

constrain the EL2 during the simulation. An MD simulation was performed for 100 ps

using the default settings of the MD routine within SYBYL and maintaining all residues

except the EL2 (Val195 to Asp223) as an aggregate to observe the behavior of the loop.

An average KOR structure was then generated from individual MD conformations and

was subsequently energy minimized. GOLD version 2.2 (166, 167) was used to dock

36 salvinorin A and 2-thiosalvinorin B into the MD-averaged and minimized KOR model and into the minimized MD snapshot model. Prior to docking, the CONCORD routine within SYBYL was used to assign initial conformations to salvinorin A and 2- thiosalvinorin B. A large radius of 25.0 Å about the Asp138 α- atom was used to define the receptor site, thus allowing for the possibility of direct interaction of salvinorin A with distant residues, including Y3137.36. The ligand-receptor complex

was energy minimized to identify and resolve any remaining strain within the system.

2.2.4.2 KOR modeling for SCAM (7)

The hKOR model presented here was built using the coordinates of the recently obtained activated bovine rhodopsin crystal (‘B’ chain of PDB code = 2I37) (23) as the initial

template. A loop search was performed to replace the missing three residues (A235 to

Q237) in the IL3 of rhodopsin, and the IL3 (K231 to A241) was subsequently energy-

minimized using the Tripos Force Field (Gasteiger-Hückel charges, distance-dependent

dielectric constant = 4.0, nonbonded cutoff = 8 Å). The N- and C-termini of the hKOR

(M1 to E33 and C322 to P327) were removed. Residues in the structurally conserved

transmembrane helical and IL1 regions were mutated to their cognate residues in the

hKOR. The rotated extracellular portion of TM2 was incorporated into the model by

replacing residues L842.51 to Y1022.69 in bovine rhodopsin with residues A1062.51 to

S1232.68 from a previously described ‘TM2-rotated’ hKOR-salvinorin A interaction model(168). The importance of Q115 as a probable H-bonding interaction site for

salvinorin A was incorporated into this ‘TM2-rotated’ hKOR model. Loop searches were

then performed to replace the remaining bovine rhodopsin segments in the model with

37 hKOR segments. The sequence was renumbered and SCWRL(169) was used to place the sidechains onto the hKOR model backbone.

In order to reproduce the characteristic features of the binding site in the previously described hKOR model(168) in which the agonist salvinorin A was bound, additional refinement of the model was carried out. First, in order to accommodate the ligand in the binding site, the EL2 was raised out of the binding cavity, enlarging it. This was accomplished by replacing the EL2 (S192 to W221) with the one from the previously described hKOR model(2, 168) in which molecular dynamics (MD) was used to enlarge the binding site cavity. Additionally, in order to reorient Y3137.36 and Y3207.43 for

consistency with the previously-proposed salvinorin A-hKOR interaction model, the

coordinates of the in the extracellular portion of TM7 (L3097.32 to N3227.45) were replaced with the coordinates of the corresponding residues in the earlier hKOR model(168). Next, the sidechain conformations of key residues in the hKOR were modified to match those in the rhodopsin template, as these sidechains had been assigned conformations by SCWRL that differed significantly from those of the original bovine rhodopsin template. As a final step in the refinement of the hKOR, the agonist salvinorin

A was placed into the binding site in a manner previously described (2), and the receptor- ligand complex energy-minimized without constraints. The stereochemical integrity of the final hKOR receptor model was verified by PROCHECK and the ProTable facility within SYBYL 7.3.

38

2.2.5 Large Quantity of KOR Expression and Purification

2.2.5.1 KOR expression

HEK293 T cells were grown in 15 cm culture dishes in medium with 10% FBS in a

humidified atmosphere consisting of 5% CO2 at 37 °C. To obtain large quantity of KOR

protein for MS, cells were transfected with the WT KOR — pcDNA3.1(+) FLAG-KOR-

His6 (25 µg/15 cm dish), using Fugene 6 transfection reagent (Roche). This pcDNA3.1(+)

FLAG-KOR-His6 vector was made by Dr. Timothy A. Vortherms, a former postdoc in

our lab. It contains an N-terminal FLAG tag and a C-terminal His6 tag. The FLAG and

His6 tagged KOR was used for RB-64 labeling and subsequent mass spectrometric

analysis. Another CHO cell line stably expressing hKOR without any tags was used for

quantifying the RB-64 (a salvinorin A analog showed covalently-bound potential) and

other drugs’ labeling ability. This CHO cell line was made by Joe Rittiner, a graduate

student in our lab. The expression of KOR in these cells has been characterized by

various radioligand binding. Our current efforts are making stable Flp-in HEK293 cell

lines (Invitrogen) expressing WT FLAG-KOR-His6 and mutants, which allow receptor

proteins expression at a consistent level.

2.2.5.2 Labeling KOR with RB-64 or RB-48

After a total of 48 hours of transfection, cell media was removed and the transfected cells

were washed by cold PBS, then the cells were labeled with RB-64 or RB-48 (5ml-10

µM/3hr for 10cm dish, 25 ml-20 µM/10hr for 15cm dish) in ice-cold PBS. Cells were

39 detached and centrifuged. Cell pellets were washed with standard binding buffer twice centrifuged at 15,000g for 20 min at 4 °C.

2.2.5.3 Anti-FLAG M2 affinity purification

The above cell pellets were resuspended in 1 ml TTSEC lysis buffer (50 mM Tris-HCl,

150 mM NaCl, 5 mM EDTA, and 2% Triton X-100, pH 7.4, one of protease inhibitor cocktail was freshly dissolved in 25 ml buffer). Membrane solubilization was completed by gently shaking and incubating on ice for 30 min. The membrane solution was centrifuged at 4 °C for 30 min at 15,000g to obtain a clear supernatant. The supernatant was collected and mixed with pre-washed and equilibrated anti-FLAG M2 affinity resin (#F2426, Sigma). For every two 15 cm dishes’ supernatant, 60 µl of 50%

M2 affinity resin slurry was added. After 1 h of rotation at 4 °C, the suspension was centrifuged at 4 °C for 1 min at 800g, and the supernatant was removed. The resin was washed four times with TTSEC buffer and centrifuged in the same conditions as above.

Then the resin was washed once again with TTSEC B buffer (50 mM Tris-HCl, 150 mM

NaCl, 2% Triton X-100, pH 7.4). Following the final wash, KOR proteins were eluted with 100 µl of 200 µM FLAG peptide (#F3290, Sigma) in TTSEC B buffer for every two

15 cm dishes’ supernatant.

2.2.5.4 Ni2+ affinity purification of KOR

The above eluted supernatant was adjusted to contain 10 mM imidazole. And for every

two 15 cm dishes’ supernatant, 20 ul of pre-washed, equilibrated Ni2+-NTA Superflow

40 resin (Qiagen) was added. The resin was pre-washed four times with 1 ml of TTSEC B buffer containing 20 mM imidazole. The mixture was incubated for 2 hr at 4 °C with gentle agitation and centrifuged at room temperature for 1 min at 800g, and then the supernatant was removed. The resin was washed six times with 1ml TTSEC B buffer.

His6-tagged KOR was eluted twice from the resin with 25 ul of TTSEC B containing 250 mM imidazole for 1 min at 4 °C. The elutions were combined in a 1.5 ml eppendorf tube and concentrated to about 10% of the original volume using Microncon centrifugal filter tubes (Millipore) with MW 10 kDa cutoff. The eluted KOR was collected and stocked in low retention tube.

2.2.5.5 SDS-PAGE and western blot analysis

Samples were treated in Laemmli sample buffer containing 1% β-mercaptoethanol at 65

°C with for 8 min. KOR proteins were resolved on 10% Tris-HCl gel. Proteins were

transferred to nitrocellulose membrane (Bio-Rad), and the membranes were prepared for

immunodetection following the general western blotting protocol. After 1 hr blocking in

milk, the membrane was incubated for 1 hr at room temperature with corresponding

primary antibodies (1:1000 dilution for anti-FLAG (#F7425, Sigma) and 1:250 for anti-

KOR antibody (#44-302G, Biosource)). After washing three times with TSET buffer, the

membranes were incubated in a 1:1500 dilution of peroxidase labeled anti-rabbit IgG-

HRP (#PI1000, Vector). Protein bands were visualized using the Supersignal west pico

chemiluminescent substrate (Thermo).

41

2.2.5.6 SDS-PAGE and Coomassie staining

Similar as the above SDS-PAGE procedure, concentrated KOR sample was treated in

Laemmli sample buffer containing 1% β-mercaptoethanol and incubated at 65 °C for 8 min. KOR proteins were resolved on 4-20% Tris-Glycine gel (Invitrogen). The SDS-

PAGE gel was stained with 0.12% R-250 Coomassie (Bio-Rad) in 10% acetic acid for 15 min on an orbital shaker at 40 rpm and destained with 10% acetic acid. The gel was saved in 1% acetic acid in a sealed plastic bag, and submitted to the proteomics center at UNC,

Chapel Hill. All the above procedure should be done in a clean hood to avoid airborne contamination.

2.2.6 Mass Spectrometric Analysis

This part of the work was performed in collaboration with the UNC-Duke Michael

Hooker Proteomics Center at UNC, Chapel Hill following their stardard protocols

(proteomics.unc.edu/protocol.shtml).

A brief introduction of MALDI-MS is provided here. Matrix assisted laster

desorption/ionization (MALDI) and electrospray (ESI) are two soft ionization techniques

that can ionize low volatility, high molecular weight molecules, such as proteins and

peptides. MALDI was invented by two teams independently in 1988 (170). The organic

compound, so called matrix, was used to absorb energy from a laser pulse and assist the

ionization of big molecules (Equation 2.3). Proteins and peptides are ionized and detected

predominantly in form — MH+. The ultraviolet or infrared laser requires matrix

42 molecules to have high UV or IR absorbance, such as benzoic and derivatives. MALDI is often combined with a time-of-flight (TOF) mass analyzer. Ions are accelerated in an electric potential field, where the velocity of an ion is proportional to its (m/z)-1/2. A detector at the end of the flight tube will produce a signal when the ion

strikes. The arriving time can be used to calculate the ion’s m/z ratio.

(ା (2.3ܪܯሻି ൅ܪݔെ݅ݎݐܽܯ՜ሺܯ൅ כݔ݅ݎݐܽܯ ,כݔ݅ݎݐܽܯ ൅ ݄߭ ՜ ܿ݅ݎݐܽܯ where M represents protein or peptide.

The detected peptide mass can be used to map the protein sequence and further determine the protein identification, so called “bottom up” MS. Peptide mass fingerprinting (PMF) is a technique to identify an unknown protein by matching the observed fragment peptide masses to the theoretical peptide masses generated from a protein database. The popular protein sequence databases include the National Center for Biotechnology Information

(NCBI) and Swiss-prot database (Swiss-Prot). Mascot (www.matrixscience.com) is a search engine for protein identifications. PMF has been successfully used for post- translational modifications. Post-translational modification widely exists in biological systems, including , phosphorylation, prenylation, acylation and .

In our study the modification is caused by covalently-bound-ligand labeling, a significant mass shift should be observed by mass analyzer.

2.2.7 Prepulse Inhibition Animal Study

This part of the work was collaborating with Dr. William C. Wetsel (Duke University).

43

Mice — Adult naïve male and female C57BL/6J mice (Jackson Labs, Bar Harbor, ME) were used in all experiments. Animals were maintained under a 14:10 hr light/dark cycle in a humidity- and temperature-controlled room with water and laboratory chow supplied.

All experiments were conducted in accordance with NIH guidelines and under an approved protocol from the Institutional Animal Care and Use Committee at Duke

University.

Salvinorin A and RB-64 Disrupt Prepulse Inhibition (PPI) in C57BL/6J Mice — Mice were administered vehicle (10% Tween 80 in Milli-Q water; 8 ml/kg, i.p.) or one of four doses of salvinorin A(0.25, 0.5, 1.0, 2.0 mg/kg, i.p.) or RB-64 (0.005, 0.01, 0.05, 0.1 mg/kg, i.p.) immediately before placement into the PPI apparatus (Med-Associates).

After 5 min acclimatization to 62 dB white-noise, animals were administered 84 test trials, beginning and ending with 10 trials each of startle-only stimuli (40 msec 120 dB white-noise burst). The remaining 64 trials were randomized between the following trial types: 8 startle-only trails, 8 trials without any stimuli (null trials), 16 trials with prepulse stimuli (4, 8, 12, and 16 dB above the 62 dB background, 4 of each intensity, 20 msec in length) that were not paired with startle stimuli (prepulse-only trials), and 32 trials of prepulse stimuli (4, 8, 12, and 16 dB above the 62 dB background, 8 at each intensity) paired with the 120 dB startle stimulus given 100 msec following the onset of the prepulse stimulus. Trials were separated by a variable interval (8-15 sec) and total test- time lasted 26-30 min for each animal.

44

Statistical analysis — The data were analyzed with SPSS 11 programs (SPSS Inc) and are presented as means ± standard error of the mean. Differences in treatment effects on null activity, baseline startle responses, and overall PPI were analyzed with ANOVA, with the main effects of dose nested within compound. Repeated measures ANOVA

(RMANOVA) was used to examine the effects of salvinorin A and RB-64 on prepulse- dependent PPI, with prepulse intensity (4, 8, 12, and 16 dB) as the within subjects effect, and compound and dose as the between subjects effects (dose nested within compound).

Differences between treatment groups were determined with Bonferroni corrected pair- wise comparisons. In all cases, p < 0.05 was considered statistically significant.

45

Chapter 3: The Binding Site of Salvinorin A (Modified from Ref(2))

3.1 Introduction and Rationale

Salvinorin A is the only known non-nitrogenous opioid receptor agonist which is highly selective for KOR and has no significant activity at DOR, MOR or ORL1, nor other tested GPCRs, neurotransmitter transporters, or ion channels (1). Because of its unique structure and selectivity for a single Class A GPCR, the salvinorin A-KOR complex provides a model system for exploring the atomic features responsible for small-molecule selectivity among highly homologous receptors. However, the binding site of KOR is not well understood even though KOR has been identified as the molecular target of salvinorin A. Our primary goal for salvinorin A study is to identify its binding site in

KOR.

Combined molecular modeling/mutagenesis studies can provide testable models for ligand interactions and activation mechanisms (171, 172) and these approaches have been used successfully for biogenic (173, 174) and peptide receptors (1). However, this combined method seems to work more successfully for ligands (either small molecule or peptidergic) which have ionic and -bond-type interactions with conserved charged residues (e.g. Asp, Glu) or residues capable of forming hydrogen bonds (e.g.

Tyr, Arg) for anchoring and orienting ligands in the binding pocket (see ref (171) for reviews). Since salvinorin A possesses no ionizable groups, ionic interactions can not provide stability in the binding pocket, although it is conceivable that hydrogen-bond and hydrophobic interactions could be involved. Salvinorin A is a triangle-like molecule with

46 three rotatable functional groups extending out its backbone, acetoxy group on C-2 position, methyl ester on C-4 position and furan group on C-12 position (Figure 3.1). Our initial hypothesis is that due to the hydrophobic nature of salvinorin A, all three functional groups of salvinorin A can interact with KOR residues through van der Waals’ forces and/or hydrogen-bonding. It is noticeable that all three groups are hydrophobic and contain a hydrogen acceptor — oxygen. To test this hypothesis, we designed mutants of

KOR which lose either van der Waals’ forces or hydrogen bonds, or both (Table 3.1). By this means, we can separate van der Waals’ forces from hydrogen bonds, and have a clear view of how salvinorin A binds to KOR.

H O Y119 (TM2)

O H O Y320 (TM7)

OH 12 ? O 11 O Q115 (TM2) H H 17 Y313 (TM7) O 9 H3C 1 2 10 8 O 35 7 O 4 6 ? L212 & F214 (EL2)

O OCH3 I294 (TM6) E297 (TM6)

Figure 3.1 The pharmacophore of salvinorin A. [Modified from Figure 5 in ref (2).]

However, the limitation of mutagenesis is that the mutation itself may cause significant structural change of the receptor and affect the binding affinity, so other biophysical methods are required to validate the mutagenesis results. In order to fully understand the

47 mechanisms for how salvinorin A binds to and activates KOR, the research process will take several steps: 1) analyzing salvinorin A’s pharmacophore, and using computer modeling to predict the key residues that are believed to contribute in salvinorin A binding to KOR (i.e. Tyr313, Gln115, Tyr 312 and Tyr139); 2) make mutations of these residues and determine if these mutations affect salvinorin A binding and function by competition radioligand binding and functional studies (e.g. Ca2+ mobilization); 3)

predict the orientation of salvinorin A in the binding site by examining specific functional

groups’ interactions with KOR, such as the interaction between substituted Cys and the

free sulfhydryl group of a salvinorin A analog; 4) use all the experimental data as

constraints to refine the docking/modeling and predict a comprehensive binding site; 5)

re-examine KOR binding site through extensive salvinorin A’s SAR study; 6) re-examine

KOR binding site by comparing with other structural studies, like X-ray crystallography,

NMR, spin labeling and SCAM (will be discussed in Chapter 4); 7) Based on all the

information obtained, novel salvinorin A analogs with special properties could be

designed (as discussed in Chapter 5).

In this chapter, we will focus on the above steps 1-5. All the mutations of KOR residues

made to identify the binding site are summarized in Table 3.1, including unpublished data

and data from recent papers (for Ki values see Table 3.2 and Table 4.6).

Table 3.1 Key residues in salvinorin A’s binding site

Observed Effects Binding Amino Hypothesized a Mutation On Binding Refb Site Acid Effects Sal A U69 Dyn A Residue? Y139 A Decrease of Minor Minor NO (2) NO

48

hydrophobic interaction and loss of hydrogen bond loss of hydrogen F Medium Medium Medium bond Decrease of hydrophobic Y312 A interaction and Minor Minor NO

loss of hydrogen (2) NO

bond

loss of hydrogen F Minor Medium NO bond Decrease of hydrophobic A interaction and Strong Medium Minor Y313 loss of hydrogen (2) Yes bond loss of hydrogen F Minor Minor NO bond Decrease of hydrophobic A interaction and Strong Medium Minor Y119 loss of hydrogen (2) Yes bond loss of hydrogen F Medium Medium Medium bond Decrease of hydrophobic A interaction and Strong Strong Minor Y320 loss of hydrogen (2) Yes bond loss of hydrogen Very F Strong Medium bond Strong Decrease of hydrophobic A interaction and Minor Minor Minor Y66 loss of hydrogen — NO bond loss of hydrogen F Minor Medium Minor bond Loss of ionic M Minor Minor Minor interaction E209 Disrupt of ionic (2) NO Q interaction and Minor Minor Minor hydrogen bond

49

Adjust the I294c A hydrophobic Minor (7) Yes

contact surface Eliminate hydrogen bond or I297 C NO (7) NO ion-ion interaction Eliminate S211 A NO (7) NO hydrogen bond Adjust the L212 A hydrophobic Strong (7) Yes

contact surface Eliminate Q213 A NO (7) NO hydrogen bond Adjust the Very F214 A hydrophobic (7) Yes strong contact surface Eliminate Q115 A Strong (7) Yes hydrogen bond a Very strong effect Ki (mutant)/Ki (wild-type) < 0.05; Strong effect 0.05 ≤ Ki (mutant)/Ki (wild-type) < 0.1; Medium effect 0.1 ≤ Ki (mutant)/Ki (wild-type) < 0.2; Minor effect 0.2 ≤ Ki (mutant)/Ki (wild-type) < 0.5; NO effect 0.5 ≤ Ki (mutant)/Ki (wild-type) < 2; Minor effect 2 ≤ Ki (mutant)/Ki (wild-type) < 5; Medium effect 5 ≤ Ki (mutant)/Ki (wild-type)<10; Strong effect10 ≤ Ki (mutant)/Ki (wild-type) < 20; Very strong effect Ki (mutant)/Ki (wild-type) ≥ 20. b — means unpublished data. c Whether I294 is in the binding site can’t be decided by the mutagenesis data, but later computer modeling, SCAM (Chapter 4) and covalent bound ligand labeling (Chapter 5) confirmed it is in binding site.

3.2 Results

3.2.1 Receptor-based Binding Site Study

3.2.1.1 Key residues in the binding site were identified by a combined

mutagenesis/computer-modeling method

We performed site-directed mutagenesis studies and characterized the binding and

functional parameters of the mutated KORs. All of the mutants were expressed at high

levels in HEK293 T cells and all had comparable affinities for the antagonist

[3H]diprenorphine (Table 3.2) and natural agonist dynorphin A (1-13), with the

50 exceptions of the Y139A (8-fold attenuation), Y119F and Y320A (6-fold attenuation each) and Y119A (4-fold attenuation). These results demonstrate that the mutations studied did not dramatically alter KOR receptor expression nor significantly perturb the topology of the receptor since binding for the high-affinity antagonist and endogenous agonist dynorphin were not greatly attenuated (e.g. < 10-fold change in affinities).

Table 3.2 Affinity (Ki, nM) of salvinorin A, U69593 and dynorphin A (1-13) binding to the WT KOR and mutants transiently expressed in HEK293 T cells

b c Ki (nM) ratio

a a Kd (nM) Bmax (pmol/mg) Sal A U69 Dyn A

Wild Type 0.25 5.8 31.6±6.5 11.7±3.4 1.75±0.35 Y139A 0.26 8.3 53.9±13.9 2 20.6±4.0 2 1.98±0.41 1 Y139F 0.59 5.0 193±38 6 101±29 9 13.3±4.6 8 Y312A 0.55 2.7 88.6±10.9 3 52.6±11.3 4 1.79±0.28 1 Y312F 0.18 4.4 65.1±11.0 2 53.0±12.5 5 1.39±0.22 1 Y313A 0.72 14.9 694±106 22 107±32 9 4.10±0.78 2 Y313F 0.26 7.0 63.3±15.2 2 37.0±5.0 3 0.93±0.14 1 Y119A 0.19 1.41 342±40 11 59.9±2.4 5 6.71±0.89 4 Y119F 0.32 3.6 233±66 7 90.7±5.3 8 11.3±4.8 6 Y320A 0.92 1.3 380±103 12 195±34 17 3.15±0.82 2 Y320F 0.82 1.7 301±75 10 276±88 24 9.68±2.84 6 a Saturation binding of 3H-Diprenorphine to the wild type and the mutants was performed of two or three independent experiments. Data represent in Mean without ±SEM, modified from table 2 of Ref(2). b The affinity constants (Ki) of the different compounds were determined in competition binding assays with 3H-diprenorphine and increasing concentrations of unlabeled compounds. Each value is the mean of three or four independent experiments (see Chapter 2 for details). c For each compound ratio is Ki (mutant)/Ki (wild-type).

51

In four out of ten of the studied KOR mutants, the affinity for salvinorin A was not significantly altered from WT KOR (Table 3.2). However, a dramatic decrease in the affinity of salvinorin A was obtained by introducing a single mutation, Y313A on TM7

(Ki = 694 nM, 22-fold decrease) — a smaller effect was seen for U69369 binding and no

significant effect was seen for dynorphin A (1-13). To elucidate the molecular

mechanism(s) responsible for salvinorin A’s selective interaction with Tyr313, we

examined a Y313F mutation which should maintain hydrophobic interactions and abolish

hydrogen-bond-type interactions between the –OH of Tyr313 and salvinorin A.

Surprisingly the Y313F mutation caused no loss of the binding affinity, indicating that

hydrophobic interactions between salvinorin A and Tyr313 provide stabilization of

salvinorin A in the binding pocket.

Of the other tested mutations, the mutations at the Tyr119 and Tyr320 loci had significant

effects while mutations at the Tyr139 locus had modest effects (<10-fold change) on

salvinorin A’s affinity for KOR. It appears that the primary mode of interaction at Tyr119

and Tyr320 is via H-bonding: in each case the majority of the affinity was lost on the

mutation from Tyr to Phe; little additional perturbation resulted from the Tyr to Ala

mutation. The affinities of U69593 and dynorphin A (1-13) were, in general, only

modestly affected by the various mutations, with exception of Y320A and Y320F.

Interestingly, the single mutation which had the greatest effect on salvinorin A’s affinity

(Y313A) attenuated U69593’s affinity 9-fold and had an insignificant effect on

dynorphin A (1-13) affinity (Table 3.2).

52

We next determined the agonist potencies and efficacies of salvinorin A, U69593 and dynorphin A (1-13) at wild-type and mutant KORs (Table 3.3). As expected, the Y313A mutation significantly attenuated salvinorin A’s potency to activate KOR (6-fold), while the Y119A mutation decreased agonist potency 9-fold. Surprisingly, several other mutations attenuated salvinorin A’s potency for activating KORs: Y139A (5-fold) and

Y312A (4-fold). These results were intriguing since Y139A and Y312A did not affect binding of salvinorin A to the KOR. The most severe effect was found by mutations

Y320A and Y320F. They abolished agonist-induced activation of KOR by all three tested compounds: salvinorin A, U69593 and dynorphin A (1-13). Aside from the Y320 mutations, Y139A (3-fold) and Y119A (4-fold), dynorphin A (1-13) tolerated mutations without losing agonist efficacy. U69593 showed a dramatic loss of potency for activation for the Y312A (19-fold) and Y119A (15-fold) mutations. As with the other compounds,

U69593’s agonist potency was diminished 6-fold by Y139A. In contrast to salvinorin A

(6-fold), Y313A only has a 2-fold effect on U69593’s ability to activate KOR.

a a Table 3.3 Agonist potency (EC50, nM) and relative agonist efficacy (normalized Emax) of salvinorin A, U69593 and dynorphin A (1-13) for the WT KOR and mutants transiently expressed in HEK293 cells.

c c EC50 (nM) ratio Emax EC50 (nM) ratio Relative Emax KOR SalA U69 DynA Wild Type 45.8±8.1 3232±728 23.6±6.4 1.13±0.19 343±54 1.53±0.26 Y139A 240±44 5 4893±795 140±38 6 1.13±0.29 1044±258 3 1.07±0.15 Y139F 47.0±22.2 1 5265±952 13.3±3.6 1 1.00±0.15 369±98 1 1.17±0.18 Y312A 189±46 4 4667±589 451±225 19 0.93±0.12 416±86 1 0.87±0.18 Y312F 68.3±22.1 1 4308±526 98.0±33.4 4 1.03±0.09 172±39 1 1.30±0.12

53

Y313A 275±53 6 5124±705 43.9±3.3 2 1.17±0.12 233±44 1 1.23±0.19 Y313F 49.6±15.6 1 5127±747 48.9±12.1 2 0.97±0.42 91±3.4 0.3 1.17±0.52 Y119A 434±163 9 5004±718 358±181 15 0.80±0.26 1290±634 4 0.80±0.26 Y119F 43.5±14.7 1 3880±770 35.9±13.5 2 0.80±0.15 276±99 1 1.03±0.03 Y320A NDb ND ND ND ND ND Y320F ND ND ND ND ND ND a EC50 and relative Emax were determined from calcium flux assays as described under Methods in Chapter 2. The results represent the average of three independent experiments with normalized Emax values in Mean ± SEM form. Modified from Table 3 of Ref(2). bND=no detectible agonist activity; mutants Y320A and Y320F have no response to salvinorin A, U69593 or dynorphin A (1-13). c For each compound ratio is EC50(mutant)/EC50(wild-type).

3.2.1.2 Specific ligand-receptor interactions determine the orientation of salvinorin

A in the binding site

To fully elucidate the orientation of salvinorin A, we examined the binding of 2-thiol salvinorin B at wt and Cys-substituted KOR mutants. We reasoned that Cys-substitutions should show enhanced affinity for 2-thionylsalvinorin if the Cys residue was in close proximity to the –SH moiety of 2-thiosalvinorin B. Since the KOR has a single Cys in the binding pocket (Cys315; (175)) we initially characterized C315S. As shown in Table 5, the C315S did not significantly alter the binding of either salvinorin A or salvinorinyl-2- thiol. Based on the salvinorin A binding model, we constructed several Cys substitutions in the background of C315S including Y119C, I294C, E297C, L309C, S310C and

Y313C (Table 3.4) of which Ile294, Glu297, Leu309, Ser310 and Tyr313 are unique to

KORs. As predicted by the salvinorin A binding model, the Y313C mutation preserved the affinity of 2-thiosalvinorin B because the –SH group is predicted to be in close proximity to the Cys substitution at the Y313 locus in the C315S background. In contrast, the affinity of salvinorin A for the Y313C-C315S double mutant decreased by 15.8-fold,

54 presumably because the 2-acetoxy group of salvinorin A cannot form interactions with

C313 as effectively or as strongly as can 2-thiosalvinorin B. Additionally, I294 and E297, when mutated to Cys, gave rise to significantly enhanced affinities for both salvinorin A and 2-thiosalvinorin B.

Table 3.4 Effect of Cys-substitution mutationsa on salvinorin A and 2-thiosalvinorin B binding to KORs. Salvinorin A 2-thiosalvinorin B KOR b c b c d Ki (nM) ratio Ki (nM) ratio ratio

Wild Type 61 ± 23 161 ± 59 2.6 C315S 38 ± 18 202 ± 72 5.4 C315S-Y313C 596 ±41 15.8 261 ± 53 1.3 0.4 C315S-Y119C 178 ± 54 4.7 226 ± 49.5 1.1 1.3 C315S-I294C 8 ± 4 0.2 45 ± 5.8 0.2 5.6 C315S-E297C 2.5 0.1 37 0.2 14.8 C315S-L309C 16 ± 14 0.4 87 ± 66 0.4 5.4 C315S-S310C 49 1.3 261 1.3 5.3 C315S-Y320C NAe NAe C315S-Y66C 85 ± 20 2.2 240 ± 25 1.2 2.8 a The double mutants were based on C315 (7.38), which is conserved among the opioid receptor family and is differentially accessible (Ref(176)). Modified from Table 4 of Ref(2). b The affinity constants (Ki) of the different compounds were determined in competition binding assays with 3H-diprenorphine and increasing concentrations of unlabeled compounds. Each value is the mean of three or two independent experiments (see Experimental Section for details). c For each compound ratio is Ki (mutant)/Ki (C315S). d For each compound ratio is Ki (salvinorin A)/Ki (2-thiosalvinorin B). e NA=NO affinity

3.2.1.3 Molecular modeling to visualize and predict the binding site

55

By docking techniques (Chapter 2 for details), the complete binding site of salvinorin A was predicted. These studies were performed in collaboration with Dr. Richard

Westkaemper (Virginia Commonwealth University) (2). The closest sites of interaction are between salvinorin A and TM1–3, TM6, TM7 and the extracellular loop 2 of the receptor model. In our model, the 2-acetoxy group of salvinorin A is positioned between the EL2 loop and the top of TM7 near Tyr313 and the furan ring hydrogen-bonds with

Tyr320 in TM7 and Tyr119 in TM2. The 4-methyl ester group is oriented toward the top of TM6. All of the mutated residues in the current model point approximately toward the central cavity, with the exceptions of Leu309 and Ser210 while Tyr312 and Tyr313 are located more remotely near the top of the binding cavity. Table 3.5 lists the distances between heavy atoms of salvinorin A and nearby amino acid residues in the KOR.

Mutations of some of the residues within a distance of 4 Å have been shown by others to affect the binding of conventional KOR ligands. These include Asp138, the putative ammonium-binding residue (177); Cys210, the EL2 loop disulfide-forming Cys (178) and Ile294, which has been implicated in a SCAM study (175). While present in the binding cavity, Tyr139 was not predicted to interact significantly with docked ligand.

Table 3.5 Residues within specified distances between the salvinorin A and the KOR receptor for the proposed binding mode. Receptor atoms may be part of the backbone or the side chain.*

3 .0Å 3.5 Å 4.0 Å 4.5 Å 5.0 Å NONE I135 Y66 M112 Y139 D138 Y119 A298 P215 E209 V134 C210 S211 I294 L212 Y313 F214

56

I316 E297 Y320 Y312 * Highlighted residues have been validated by mutagenesis study (table 3.1). [Modified from Table 1 of Ref(2)].

Figure 3.2 Modeling salvinorin A-KOR interactions. The proposed binding mode of salvinorin A in the KOR was looking through the extracellular side of the helical bundle. Tyr313 and Tyr320 in TM7 are nearest to the viewer. The KOR is color-coded based on assignment of secondary structure using the Kabsch-Sander (36) algorithm (red=helix, blue=, violet=turn, yellow=coil). The ligand and important nearby residues are rendered as capped stick figures. [Modified from Figure 5c from ref(2).]

3.2.2 Ligand-based Binding Site Study

This part of the work was done by collaborating with two medicinal chemistry labs, Dr.

57

Jordan K. Zjawiony (University of Mississipi) and Dr. Mark A. Rizzacasa (The

University of Melbourne, Australia). In this dissertation, I only present the data of the salvinorin A derivatives that have been synthesized by our collaborators and tested in Dr.

Bryan Roth’s lab; for details please refer to the refs (179, 180). The covalently-bound salvinorin A compounds will be discussed in chapter 5.

This extensive SAR study was to identify key functional groups in the salvinorin A structure. All salvinorin A analogs were screened in competition radioligand binding assays with cloned opioid receptors (Table 3.6). Compounds with submicromolar affinity were also screened for functional activities using calcium mobilization (or Ca2+ flux) assay. The results suggest that the acetoxy group on C-2 position, methyl ester on C-4 and furan ring on C-12 are required for salvinorin A binding, while the lactone (C-17) and ketone (C-1) functionalities are not (5, 159). For all the isolated from S.

Divinorum, salvinorin A appears to be the only one to have activity at KOR, whereas other salvinorins show negligible affinity. Extending the alkyl chain and adding bulky aromatic groups to C-2 were found to diminish, if not totally abolish, the affinity for

KOR(159). The acetoxy group is therefore the optimal alkyl chain length. Bioisosteres of salvinorin A were also developed and evaluated in binding and functional assays.

Replacement of the oxygen with sulfur on C-2 produced the most similar analog of salvinorin A. The C-2 thioacetate isostere indeed produced comparable activity to salvinorin A, but nitrogen substitution had a diminishing effect. Intermediates, which lack a β – carbonyl at C-2 displayed moderate affinity (6). The difference in affinity of sulfur and nitrogen may be attributed to the presence of two lone pairs of electrons, just like

58 salvinorin A itself, in thioacetate which may play a role in receptor recognition. A hydrogen bond acceptor at C-2 may facilitate high affinity binding, but is not critical in receptor recognition when considering the affinities of intermediates. In fact, a C-2 methoxymethyl analog of salvinorin A was reported to have a 7-fold increase in potency for KOR using [3H]diprenorphine as the radioligand (179). It has also been

shown that replacing the acetate group with a formate group lowers KOR affinity (5).

The stereoisomer derivatives of salvinorin A were also examined (4, 5). Sulfur analogs of

salvinorin A and B with natural (α) and inverted (β) configurations at carbon C-2 were

designed. Thiolanalogs with the same configuration at C-2 as in natural salvinorin A

showed higher affinity to KOR than their corresponding epimers. The novel class of

salvinorin A derivatives, hemiacetal, were recently tested. The dimethyl ester derivative

of the hemiacetal was found to bind to both KOR and MOR with a 9-fold selectivity for

KOR (3). However, after , compound 52 is practically devoid of affinity

(Table 3.6).

Table 3.6 Salvinorin A derivatives tested in Dr. Bryan Roth’s lab.* Structure Binding affinity Functional activity No. Derivative Name Ki, nM Ec50, nM, Emax, (Receptor, Hot ligand, (Receptor, Assay, Ref) Molecular Weight Reference) O

4 ± 1 O O (rKOR, [3H]U69,(5)) HH 46 ± 8 RO 39.6 ± 14.5 O 100 % (hKOR, [3H]Dipre,(2)) (hKOR, Ca2+ flux, (5)) 1 239439 3 17 ± 6 O OCH (hDOR, [ H]DADLE) 3 100% 274324 (rKOR, [35S]GTPγS) R=Ac (hMOR, [3H]DAMGO) Salvinorin A

C23H28O8 432.46

59

O

O O H H O O 163 ± 50 244 ± 102 2 O (rKOR, [3H]U69,(5)) 78 % (hKOR, Ca2+ flux, (5)) O OCH3 8-epi-salvinorin A C23H28O8 432.46

8672 3 Salvinorin B (rKOR, [3H]U69,(5)) Not active at 10 μM C21H26O7 390.43 18 ± 2 (rKOR, [3H]U69,(5)) 51.18 ± 5.42 (rKOR, [3H]U69) 315 ± 35 4 2-salvinorinylformate 20945 108 % C22H26O8 ( hDOR, [3H]U69) (hKOR, Ca2+ flux, (5)) 418.44 61027 (hMOR, [3H]U69)

O

R1O O H H R O 2 O 1022 ± 262 5 (rKOR, [3H]U69,(5)) Not active at 10 μM O OCH3 R1=R2=Ac Salvinorin C C25H30O9 474.50

R1=H, R2=Ac >10000 Salvinorin D (rKOR, [3H]U69,(5)) Not active at 10 μM 6 C23H28O8

432.46 R1=Ac, R2=H >10000 Salvinorin E (rKOR, [3H]Dipre, (5)) Not active at 10 μM 7 C23H28O8

432.46 O

R1O O H H R O >10000 2 O 8 (rKOR, [3H]U69,(5)) Not active at 10 μM

O OCH3 R1=R2=Ac

60

Dihydrosalvinorin C C25H32O9 476.52 R1=Ac, R2=H 1125 ± 365 Dihydrosalvinorin E (rKOR, [3H]U69,(5)) Not active at 10 μM 9 C23H30O8

434.48 O

O O HH O O 156 ± 18 126 ± 36 O (rKOR, [3H]U69,(5)) 108% 10 (hKOR, Ca2+ flux, (5)) O OCH3

13,14,15,16-tetrahydrosalvinorin A C23H32O8 436.50 O

O O HH O OH 59 ± 11 O 78 ± 21 (rKOR, [3H]U69,(5)) 11 107% O OCH 3 (hKOR, Ca2+ flux, (5))

17-salvinorinyl lactol (predominantly β anomer) C23H30O8 434.48 O

O O HH O 6 ± 1 223 ± 60 O (rKOR, [3H]U69,(5)) 103% 12 (hKOR, Ca2+ flux, (5)) O OCH3

17-deoxysalvinorin A C23H30O7 418.48 O

O O H O 6 ± 2 624 ± 200 O (rKOR, [3H]U69,(5)) 116% 13 (hKOR, Ca2+ flux, (5)) O OCH3 8,17-didehydro-17-deoxysalvinorin A C23H28O7 416.46

61

O

O O HH 347 ± 53 O 3 O (rKOR, [ H]U69,(5)) >10000 14 O Antagonist-like (hKOR, Ca2+ flux, (5)) OH 18-hydroxysalvinorin A C22H28O7 404.45 O

O HH O 18 ± 2 O 141 ± 43 (rKOR, [3H]U69,(5)) O 122% 15 (hKOR, Ca2+ flux, (5))

O OCH3 1-deoxysalvinorin A C23H30O7 418.48 O

O O HH 157 ± 40 HS 3 O (hKOR, [ H]Dipre, (2)) 287 ± 85 16 54.5 ± 25.7 89% (rKOR, [3H]U69, (4)) (hKOR, Ca2+ flux,(4)) O OCH3 2α-salvinorinyl thiol C21H26O6S 406.49 O

O O HH HS O 151 ± 53 123 ± 30 17 (rKOR, [3H]U69, (4)) 106% (hKOR, Ca2+ flux,(4)) O OCH3 2β-salvinorinyl thiol C21H26O6S 406.49 O

O O 18.4 ± 7.9 HH 3 S (rKOR, [ H]U69, (4)) 4.77 ±2.72 O 8 ± 1 107% O 18 (rKOR, [3H]U69,(6)) (hKOR, Ca2+ flux,(4))

O OCH3 2α-salvinorinyl thioacetate C23H28O7S 448.53

62

O

O O HH S 546 ± 140 O >2000 (rKOR, [3H]U69, (4)) O 71% 19 (hKOR, Ca2+ flux,(4))

O OCH3 2β-salvinorinyl thioacetate C23H28O7S 448.53 O

O O HH R3 O 3347 ± 1115 608 ± 103 76% (rKOR, [3H]U69,(6)) 20 (hKOR, Ca2+ flux,(6))

O OCH3

R3=Cl Salvinorinyl-2-chloride C21H25ClO6 408.87 R3= Cl3C O 375 ± 42 1012 ± 244 O (rKOR, [3H]U69,(6)) 101% 21 Salvinorinyl-2-trichloroacetate (hKOR, Ca2+ flux,(6)) C23H25Cl3O8 535.80 F3C O R3= 825 ± 204 211 ± 37 O 103% (rKOR, [3H]U69,(6)) 22 Salvinorinyl-2-trifluoroacetate (hKOR, Ca2+ flux,(6))

C23H25F3O8 486.44 261 ± 67 2704 ± 480 (rKOR, [3H]U69,(6)) 59% Salvinorinyl-2-bromide 23 (hKOR, Ca2+ flux,(6)) C21H25BrO6

453.32 H N R3= 78 ± 1 86 ± 22 87% O (rKOR, [3H]U69,(6)) 24 (hKOR, Ca2+ flux,(6)) 2-salvinorinylamide C23H29NO7 431.48 R3= N3 >4000 140 ± 28 Antagonist-like 2-salvinorinylazide (rKOR, [3H]U69,(6)) 25 (hKOR, Ca2+ flux,(6)) C21H25N3O6 415.44 O 32.63 4.7 R3= (rKOR, [3H]Bremazocine, 100% 26 O (159)) (hKOR, cAMP, (159)) Salvinorinyl-2-propoinate

63

C24H30O8 446.49

O >10000 R3= (rKOR, [3H]Bremazocine, 27 O (159)) Not active at 10 μM Salvinorinyl-2-privalate C26H34O8 474.54 R3= O O >10000 O (rKOR, [3H]Bremazocine, 28 Salvinorinyl-2- Not active at 10 μM ethylcarbonate (159)) C24H30O9 462.49 Cl3C O O R3= >10000 O (rKOR, [3H]Bremazocine, 29 Salvinorinyl-2- (159)) Not active at 10 μM triethylcarbonate C24H27Cl3O9 565.82

R3= O >10000 O (rKOR, [3H]Bremazocine, 30 Salvinorinyl-2- (159)) Not active at 10 μM cyclopanecarboxylate C25H30O8 458.50 O 3199 R3= 40 (rKOR, [3H]Bremazocine, O 34% Salvinorinyl-2-heptanote (159)) 31 (hKOR, cAMP, (159)) C28H38O8

502.60 H N > 10000 R3= (rKOR, [3H]U69) Not determined 32 n-propyl-2-salvinorinylamine C24H33NO6 431.52 O

N R3= No affinity 33 n-acetyl-n-propyl-2- (rKOR, [3H]U69) Not determined salvinorinylamine C26H35NO7 473.56 R3= H N 2 391 ± 71 2-salvinorinylamine 3 34 (rKOR, [ H]U69) Not determined C21H27NO6

389.44 R3= O 17.2 ± 3.2 O (rKOR, [3H]U69) Not determined 35 2-salvinorinyl butanoate C25H32O8 460.52

64

O R3= 103 ± 26 O 3 36 2-salvinorinyl valerate (rKOR, [ H]U69) Not determined C26H34O8 474.54 R3= O 103 ± 30 O 3 37 2-salvinorinyl hexanoate (rKOR, [ H]U69) Not determined C27H36O8 488.57

0.59 ± 0.21 (rKOR, [3H]U69) 0.077 ± 0.016 38 RB-64 39 ± 11 (rKOR, [35S]GTPγS) C24H27NO8S (rKOR, [3H]Dipre) 489.53 O R3= MeO 20 ± 1 (rKOR, [3H]U69) O 39 430 ± 50 Not determined RB-65 3 C24H30O9 (rKOR, [ H]Dipre) 462.48 Cl O R3= 22 ± 10 O (rKOR, [3H]U69) 40 α,β-chloro-propionyl 190 ± 70 Not determined -salvinorin (RB-55) (rKOR, [3H]Dipre) C24H29ClO8 480.93

O R3= NC 74 ± 34 3 O (rKOR, [ H]U69) 41 1010 ± 200 Not determined RB-59 3 C24H27NO8 (rKOR, [ H]Dipre) 457.47 Cl O 30 ± 15 R3= (rKOR, [3H]U69) 42 O 270 ± 50 Not determined RB-55-1 (rKOR, [3H]Dipre) C24H29ClO8 480.93 Cl O 58 ± 32 R3= 3 (rKOR, [ H]U69) 43 O 4290 ± 2920 Not determined RB-55-2 (rKOR, [3H]Dipre) C24H29ClO8 480.93 Cl 1970 ± 890 O 3 44 R3= Cl (rKOR, [ H]U69) Not determined

O

65

RB-66 C23H26Cl2O8 501.35 O R3= Cl 3.43 ± 1.50 X 10-5 O 2.10 ± 0.84 0.19 ± 0.01 RB-48 (rKOR, [3H]U69) 102% 45 C23H27ClO8 32 ± 15 (rKOR, [35S]GTPγS) 466.91 (rKOR, [3H]Dipre)

O R3= Br O 1.46 ± 0.86 46 RB-50 (rKOR, [3H]U69) Not determined C23H27BrO8 511.35

O R3= F O 47 KH-19 Being tested Not determined C23H27FO8 450.45

R3= Cl O 0.27 ± 0.24 O 3 LP-0511 (rKOR, [ H]U69) 48 41 ± 8 Not determined C24H29ClO8 3 480.94 (rKOR, [ H]Dipre)

O

O O O HN O 224.0 ±84.0 (rKOR, [3H]U69) Not determined 49 H2N O 2-acetylamino-4-salvinorinylamide C22H28N2O6 416.47

O

O O H N 2 O 143.3 ±26.9 50 (rKOR, [3H]U69) Not determined

H2N O 2-amino-4-salvinorinylamide C20H26N2O5 374.43

66

O

O H HO CO2H NA for KOR and MOR 51 OH (rKOR, [3H]U69 ) Not determined (hMOR, [3H]DAMGO) CO2H Hemi-acetal C20H26O8 394.42 O

O H R4O CO2Me OH 219±59 (rKOR, [3H]U69) Not determined 52 1926 ± 147 CO2Me (hMOR, [3H]DAMGO) R4 = H Diester C22H30O8 422.47

R4=Ac 6003 ± 1242

(rKOR, [3H]U69) Acetate Not determined 53 7487 ± 2141 C24H32O9 (hMOR, [3H]DAMGO) 464.51 O

O OH H HO CO Me 1991±70 2 (rKOR, [3H]U69) Not determined 54 >10000 3 CO2Me (hMOR, [ H]DAMGO) Hydroxyester C22H30O8 422.47 O

O HO H AcO CO Me 2 6000 ± 1200 (rKOR, [3H]U69) 55 Not determined CO2Me ”Acetylated hemi-ketal” (RB-58) C24H32O9 464.51

* For unpublished data, there is no reference listed.

67

3.3 Discussion

In the mutagenesis studies, we identified key residues in KORs responsible for the high

binding affinity and efficacy of salvinorin A. Surprisingly, we discovered that salvinorin

A was stabilized in the binding pocket by interactions with residues in helix 7

(Tyr313 and Tyr320) and in helix 2 (Tyr119). By contrast, the prototypical nitrogenous

KOR agonist U69593 and the endogenous peptidergic agonist dynorphin A (1-13) showed differential requirements for these three residues for binding and activation.

These results are important because they demonstrate that salvinorin A’s exquisitely potent and efficacious interactions with KORs are due to novel modes of binding within a common three-dimensional space shared by structurally diverse agonists, each of which utilizes different residues for binding and activating KORs. We also employed a novel approach whereby we examined the effects of Cys-substitution mutagenesis on the binding of salvinorin A and an analogue with a free sulfhydryl group — salvinorinyl-2- thiol. We discovered that residues predicted to be in close proximity — especially Tyr

313 — to the free thiol of salvinorinyl-2-thiol when mutated to Cys showed enhanced affinity for 2-thiosalvinorin B. Taken together, these findings imply that the diterpenoid salvinorin A utilizes unique residues within a commonly shared binding pocket to selectively activate KORs.

In the salvinorin A binding model described here, the mutated residues point toward a central cavity, although some are clearly less sterically accessible to bound ligands — particularly Tyr312 and Tyr313. Neither Y312A nor Y312F mutation affects salvinorin

A’s affinity for the KOR. This is consistent with the proposed model, which indicates

68 only a weak interaction with Tyr312. Interestingly, the Y313A mutation has a large effect on affinity (22-fold) while Y313F has no significant effect. The results of the Y313A and

Y313F mutations are consistent with Tyr313 stabilizing the ligand via a hydrophobic- type interaction and with our docked salvinorin A model which predicts a direct hydrophobic interaction with Tyr313. Though potentially accessible to small molecules,

Tyr313 was positioned in a more remote area at the top of the binding cavity interacting with residues in the EL2. Tyr313 was previously proposed to provide a hydrogen bond to the 2-acetoxy carbonyl of salvinorin A based on docking studies performed using a de novo model developed by Mosberg’s group (181). The current model that is based explicitly on the experimental rhodopsin crystal structure shows a substantial difference in the disposition of Tyr313. The fact that Y313F has no effect on ligand affinity but

Y313A reduces affinity by 22-fold has compelled us to modify our original model wherein we proposed that Tyr313 interacted with salvinorin A primarily via hydrogen- bonding type interactions. Based on our current findings, we propose that a hydrophobic interaction is more likely. In the proposed salvinorin A binding model, the 2-acetoxy group of salvinorin A provides the requisite hydrophobic interaction with Y313, so that salvinorin A retains affinity for the Y313F mutation, but loses affinity (22-fold decrease) for the Y313A mutation.

Mutations at other residues also had significant effects on the binding of salvinorin A.

Although accessible, the closest Tyr139 side chain-ligand distance is at least 4 Å away — consistent with the weak effect of either Tyr139 mutation. In our proposed binding model, it is possible for the 17-oxo group to form a weak hydrogen bond with Tyr139;

69 the donor-hydrogen-acceptor (D-H-A) angle is roughly 130° when the ligand carbonyl oxygen-Tyr139 side chain oxygen distance is 3.0 Å. However, in order for this hydrogen bond to be formed, the Tyr139 τ1 torsion angle must assume values that give rise to higher-energy eclipsed conformations, consistent with the modest 6-fold increase in the

Ki observed for the Y139F mutation. The Y139A mutation (only 2-fold decrease in

affinity) suggests that an amino acid side chain that has a small hydrophobic group, rather

than a large hydrophobic group, is beneficial when the interaction takes place with a

hydrophilic group on the ligand. Both mutations of Tyr320 result in loss of affinity for

salvinorin A by about 10-fold, suggesting hydrogen bond involvement. Our proposed

model interacts with Tyr320 via hydrogen bonding with the furanyl of

salvinorin A. The interatomic distance between the furan oxygen atom in salvinorin A

and the Tyr320 side chain oxygen is 3.0 Å. Mutation of Tyr119 to either Phe or Ala

decreases the affinity of salvinorin A for the KOR, but to a somewhat lesser extent than

for the analogous Tyr320 mutations, again indicating that hydrogen bond interactions are

involved. As with Tyr320, our proposed salvinorin A model interacts with Tyr119 via

hydrogen bonding with the furanyl substituent of the ligand (Figure 3.1). It places the

furan ring oxygen atom somewhat farther away from Tyr119 (3.6 Å) than from Tyr 320

(3.0 Å), potentially explaining the slightly greater decrease in binding affinity for the

Tyr320 mutation when compared to the analogous Tyr119 mutation.

To further explore the interactions of the salvinorins with the KOR, we employed a novel

approach whereby we combined Cys-substitution mutagenesis with an evaluation of the

binding of salvinorinyl-2-thiol and salvinorin A. We reasoned that if residues mutated to

70

Cys were in close proximity to the thiol of 2-thiosalvinorin B, there should be an enhancement (or at least a retention) of 2-thiolsalvinorin B’s affinity for the KOR, while the affinity for salvinorin A would likely decrease. An inspection of the salvinorin A binding model disclosed that Tyr313, when mutated to Cys, would yield a Cys residue that is predicted to be in close proximity to the free thiol of salvinorinyl-2-thiol. In addition, mutating other nearby residues in a similar fashion would produce Cys residues nearer to other positions on the salvinorin molecule. If our proposed model is correct, mutations at these non-Tyr313 positions should affect the binding of salvinorin A and salvinorinyl-2-thiol roughly equally, since salvinorin A and salvinorinyl-2-thiol differ only at the 2-position. The results of these experiments are presented in Table 3.4 and agree with the predictions.

In our model of the KOR (Figure 3.2), the side chain of Cys315 is located in the interface between helices TM6 and TM7, so it is not surprising that the mutation C315S did not result in a statistically significant change in the binding affinity for either salvinorin A or salvinorinyl-2-thiol (Table 3.4). The most significant result from the Cys-substitution mutation studies is that the double mutant C315S-Y313C KOR has 15.8-fold less affinity for salvinorin A than does the C315S single mutant KOR, whereas the same double mutant KOR’s affinity for 2-thiolsalvinorin B is largely unaffected compared to the single-mutant C315S KOR. This would suggest that it is indeed the 2-position of the salvinorins that interact with Tyr313, since an SH-acetoxy interaction (in the case of salvinorin A) would be very weak and would increase the Ki significantly, whereas an

SH-SH interaction (in the case of 2-thiolsalvinorin B) would be stronger due to

71 hydrophobic interactions and/or disulfide bond formation, and would allow the double mutant to retain affinity for 2-thiolsalvinorin B. We believe disulfide bond formation is unlikely since preliminary studies have demonstrated that prolonged exposure to 2- thiosalvinorin B does not lead to an irreversible loss of binding (data not shown). Our proposed model also positions the 4-methyl ester group in very close proximity to Ile294.

The double mutations C315S-I294C and C315S-E297C each affect both salvinorins roughly equally, increasing the affinity of both by 5- to 10-fold. The nature of these interactions is unclear, since both hydrophobic and hydrophilic hydrogen bond acceptor regions are present. However, visual inspection of space-filling models reveals that the terminal methyl of the ester group at the 4-position of salvinorin A can interact with hydrophobic portions of the Ile294 side chain. Mutation to Cys would retain some hydrophobic interaction potential, and perhaps more importantly, would remove some of the steric bulk in the region, allowing the 4-position to more effectively associate with the side chains at positions 294. The double mutations involving Leu309 and Ser310 did not significantly alter the binding affinity of either salvinorin, and this is consistent with our proposed model complex in that these residues are not part of the ligand binding site.

Taken together, these findings support a mode of binding whereby salvinorin A and 2- thiosalvinorin B interact with the KOR via residues that are not utilized by conventional

KOR peptide and non-peptide agonists (e.g. dynorphin A (1-13) and U69593, respectively). These residues most likely line a putative binding pocket that overlaps in three-dimensional space with that used by nitrogenous KOR agonists. Thus, compounds like U69593 are predicted to bind in approximately the same three-dimensional space but

72 do so by utilizing different residues. It is likely that salvinorin A’s extraordinary selectivity and potency for the KOR is due to the fact that it uses these unconventional and generally non-conserved residues for ligand binding. Residues with which our proposed model interacts (namely Ile294 and Tyr313) are unique to the KOR. In addition, since there is no relatively strong salt bridge anchoring salvinorin A into the binding pocket, and since there are many hydrogen bonding and lipophilic interaction sites within the pocket, it would not be unexpected for the KOR to recognize salvinorin A via more than one binding mode. It is possible that these various models, taken together, could collectively give rise to the observed affinity and/or activation of the KOR, and that no individual bound conformation would be totally responsible for the observed effects.

73

CHAPTER 4: The Conformational Changes of KOR Induced by G protein-Coupling (Modified from Ref (7))

4.1 Introduction and Rationale

Recently crystal structures of the human β2AR have been obtained using two different approaches to stabilize receptor protein and increase polar surface area (24, 25, 182).

Together with the rhodopsin structure (17, 18, 183), high-resolution GPCR X-ray

structures represent an important breakthrough in understanding the molecular

mechanism of how GPCRs function. However, to form a crystal, a GPCR has to be

locked in a single conformational state. This is a significant limitation since there is a

large amount of functional and biophysical evidence showing that GPCRs are

conformationally complex and dynamic. GPCRs are very likely to adopt conformations

specific for the bound ligand and the associated signaling protein (e.g. G proteins,

arrestins). Although ligand-induced/stabilized conformational changes in GPCRs are well

documented (184, 185), there is little direct evidence for G protein-induced

conformational changes in any GPCR. Our goal is to investigate the potential

conformational changes in KOR induced by G protein-coupling and how these

conformational changes affect agonist interactions at molecular level.

First, we needed to design a cellular system where G proteins can induce conformational

changes in KOR. It has been reported that G proteins are present in 10-100 molar excess

compared to GPCRs(186, 187), the overexpression of KOR, 2 pmol/mg as compared to

74 less than 20 fmol/mg in native brain tissue, can consume the free G proteins and leave a significant portion of KOR uncoupled (Figure 4.1). In an analogous way, the overexpression of G protein changes this ratio with the consequence that the percentage of GPCR which is ‘precoupled’ to G proteins increases(188). To investigate this possibility, the effects of over-expressing Gα subunits (Gα16 or Gαi2) with the κ-opioid

receptor (KOR) were examined (Figure 4.1). Since opioid receptors are capable of

coupling with the pertussis toxin sensitive Gα–subunits Gi/o and the pertussis toxin

insensitive Gz and Gα16 subunits(189, 190), two types of G proteins were chosen for

study: Gαi2 which has demonstrated preferential coupling to KOR and the promiscuous

Gα subunit Gα16.

Second, we need an appropriate approach to examine the conformational changes of

KOR. There are several methods have been used to interrogate GPCR conformational

changes (Table 4.1). Every method has its limitations and advantages. On the whole, the

substituted cysteine accessibility method (SCAM) fits our research goal better than the

others. Therefore the SCAM approach was used along with the sulfhydryl reagent (2-

aminoethyl)methane thio sulfonate (MTSEA) to probe conformational changes which

might occur in transmembrane domains 6 (TM6), 7 (TM7) and extracellular loop 2 (EL2)

of the KOR in various G protein backgrounds. Also, a kinetic study was applied to

confirm the conformational changes identified by SCAM.

To our knowledge, this is the first direct study investigating how G protein alpha subunits

induce different conformational changes in their cognate GPCRs. This research will help

75 us gain deeper insight into GPCR signaling and function of receptor complex, especially the intriguing phenomena of “functional selectivity”. For many years it has been documented that GPCR agonists display functional selectivity or, the ability of ligands

(both agonists and antagonists) to differentially modulate signaling pathways depending on the cellular milieu (191-195). One possible explanation for the phenomenon of functional selectivity is that G proteins differentially shift the conformation of GPCRs from the ground state to a series of activated states. Since G proteins are heterogeneously distributed in various cell compartments and show cell type and developmental expression patterns (196-198), it has been suggested that the cellular expression profile of

G proteins and effectors will affect the pattern of activation of downstream signaling pathways (199-201). According to this model, different agonist molecules preferentially sample some conformational changes over others, leading to the establishment of an agonist-preferred G protein-coupling pathway. G proteins have recently been shown to precouple with receptors specifically before the addition of agonists (187, 202) leading, perhaps, to distinct conformations. Moreover, recent reviews of the G protein-dependent pharmacology of ligands suggest the promise for designing conformationally selective ligands (203), i.e. ligands that bind preferentially to particular GPCR conformations.

KOR KOR•Gα16 KOR•Gαi2

No Gα subunit Gα -overexpressed cells Gα -overexpressed cells overexpression 16 i2

KOR Non KOR-coupling Gα subunits: Gα etc. Weakly KOR-coupling Gα subunit: Gα Strongly KOR-coupling Gα subunit: Gαi2 s 16

76

Figure 4.1 Gα subunit overexpression was used to stabilize receptor conformations in the various active states. In each cell system, KOR exists in both G protein-coupled and uncoupled forms. However, KORs are more likely to be in the G protein-coupled form when G protein alpha subunits (such as Gα16 and Gαi2) are overexpressed. For clarification, abbreviations were used to describe the three cell systems: KOR, KOR•Gα16 and KOR•Gαi2.

Table 4.1 The Comparison of Common Biochemical and Biophysical Methods Used for GPCR Conformation Study Method Mechanism Conclusion Limitation Ref Introduced Cys Nitroxide radicals labeling mutations may Cys, conformational changes Rhodopsin activation Spin change the whole detected by Electron involves TM3 and (204) structure; not much Labeling paramagnetic resonance TM6 movements. details other (EPR) individual residues Introduced Cys mutations may Fluorophore coupling to change the whole Cys, and nearby Trp structure; Full activation of quenching the fluorescence, fluorescence may be β2AR needs disrupting (205) Fluorescence conformational changes affected by native ionic lock. detected by Fluorescence aromatic residues; . not much details other individual residues Introduced Cys Methanethiosulfonate TM6 rotating/tilting mutations may modifying Cys, detected by associated with β2AR (206) SCAM change the whole Radioligand binding activation structure Introduced His, Cys Cu2+ or Zn2+ chelating with mutations may Asp, His, Cys; β2AR, TM6 and TM7 Metal Ion change the whole conformational changes move apart with large (207) structure; not much Chelating detected by functional assay amplitude. details for other ( cAMP production). individual residues Muscarinic agonists Introduced mutations N-ethylmaleimide trigger a separation may change the Disulfide crosslinking two close Cys between the whole structure; residues, conformational cytoplasmic regions crosslinking is not (208) Crosslinking changes detected by MW between TM1 and specific reaction; not shift on Western Blot. TM7, inverse agonists much details other increase the proximity. individual residues MOR undergoes Only global changes Antibody can recognize the conformational for N-terminus, no Region middle and distal part of N changes following details for TMs; terminus, conformational receptor activation, (209) Specific antigen-antibody changes detected by antigen- antibodies differentiate Antibody interaction is not antibody interaction. ligands with varying specific. efficacies. NMR 9F-, or 15N-labeling select Rhodopsin Try side Very complex (210)

77

amino acids, conformational chains prefer one protein purification changes detected nuclei specific conformation, and labeling chemical shift presented in while backbone procedure, and NMR. fluctuates on micro- to indirect readouts by millisecond time scale. complex NMR graph.

4.2 Results

4.2.1 Examine the Conformational Changes of KOR by SCAM

4.2.1.1 Characterize Cys mutants in TMs 6 and 7 and EL2 of KOR.

Prior to performing Cys-accessibility studies, a large number of Cys mutants were

created. For these studies, 23 consecutive residues in TM7 (not including C3157.38), 6 residues in the upper part of TM6 and 11 residues in EL2 of KOR were mutated to Cys

7.38 3 based on the C315 S background(211). Kd and Bmax values using [ H]diprenorphine for

TM6 and TM7 Cys mutants have been previously reported by others(157, 158). Our pattern of results was similar and showed only minor alterations in diprenorphine affinity

(data not shown). For EL2 mutants, Kd and Bmax values are summarized in Table 4.2.

Relatively minor alterations in Kd values (0.13 nM ~ 0.63 nM) were found which are

similar to wild type KOR (0.46 nM) while the Bmax ranged 0.022 to 1.1 pmol/mg (Table

4.2). These results indicate that the Cys mutagenesis does not greatly alter antagonist

binding affinity. The sole exceptions were the P3277.50C mutation which resulted in a

non-expressed receptor protein as judged by radioligand binding studies, and the

Y3207.43C mutation. Surface biotinylation and anti-FLAG immunoblotting confirmed that

the P3277.50C mutant was not expressed and that the Y3207.43C, along with other selected

mutants were expressed on the plasma membrane (data not shown). Taken together, these

findings indicate the examined Cys mutations do not drastically affect the binding pocket

78 for diprenorphine and that among the mutants evaluated surface expression is normal.

3 Table 4.2 Kd and Bmax values of [ H]diprenorphine binding to the WT KOR and EL2 Cys mutants K B Ratiob EL2 Mutantsa d max nM pmol/mg Kd(mutant)/Kd(WT) WT KOR 0.46 ± 0.11 2.7 ± 0.5 C315S-V205C 0.43 ± 0.07 1.1 ± 0.3 0.9 C315S-D206C 0.29 ± 0.05 0.61 ± 0.28 0.6 C315S-V207C 0.60 ± 0.06 1.0 ± 0.2 1.3 C315S-I208C 0.49 ± 0.10 0.74 ± 0.13 1.1 C315S-E209C 0.63 ± 0.14 0.70 ± 0.14 1.4 C315S-S211C 0.38 ± 0.11 0.83 ± 0.17 0.8 C315S-L212C 0.13 ± 0.04 0.022± 0.002 0.3 C315S-Q213C 0.48 ± 0.03 0.99 ± 0.22 1.0 C315S-F214C 0.24 ± 0.07 0.023 ± 0.002 0.5 C315S-P215C 0.36 ± 0.13 0.39 ± 0.12 0.8 C315S-D216C 0.39 ± 0.08 0.87 ± 0.31 0.8 aSaturation binding of [3H]diprenorphine to the wild type and the EL2 mutants was performed according to the procedure in Chapter 2 MATERIALS AND METHODS. Data represent in mean ±SEM from two to four independent experiments. Receptors are transiently expressed in HEK 293T cells. b The ratio is Kd(mutant)/Kd(WT).

4.2.1.2 SCAM elucidates conformational changes induced by G proteins

Co-overexpression of Gα subunits and GPCRs will likely modify the local G protein

environment sensed by GPCRs, leading to potential conformational changes in the

GPCRs. These conformational changes induced by Gα subunits can be reflected by a change in the pattern of SCAM-sensitive residues. Prior to determining potential changes in conformation, however, the basal MTSEA sensitivity patterns for the various Cys mutants needed to be determined.

79

In initial studies, 7 out of the 23 Cys mutants in TM7 of KOR were identified as being significantly more sensitive to the MTSEA reagent than the C3157.38S ‘Cys-less’ KOR as

judged by an analysis of variance in HEK 293T cells: S3107.33, F3147.37, I3167.39,

A3177.40, L3187.41, G3197.42 and Y3207.43 (Figure 4.2a). Upon stable over-expression of

7.34 7.49 Gα16, an additional two residues—S311 C and N326 C—became sensitive (Figure

4.2a). Upon stable over-expression of the ‘specific’ Gα subunit Gαi2 even more residues

became sensitive (Y3137.36, N3227.45, S3237.46 and L3297.52). In addition, the absolute

magnitude of average inhibition induced by the MTSEA reagent increased to 0.20 upon

Gαi2 overexpression (Table 4.3). In addition to a global change in TM7 residue

sensitivity, there was an interesting switch for a critical residue essential for salvinorin A

7.36 binding—Y313 C (2), which changed from being insensitive to sensitive upon Gαi2 overexpression.

For TM6, only residues which had previously been determined to be ‘sensitive’ by other investigators were tested(158). It was found that the upper part of TM6 displayed only a limited change in the absolute amount of sensitivity from 0.04 in Gα16 cells to 0.03 in

6.57 Gαi2 cells (Table 4.3). An interesting observation was that V296 C became insensitive

after Gαi2 over-expression (Figure 4.2b). These findings are consistent with a model

which implies that V2966.57 is a half-turn from I2946.55 and is facing either other TMs or

lipids. EL2 also presented a significant change in overall inhibition (0.06 for Gα16 and

0.17 for Gαi2 respectively), with two residues L212 and F214 being identified as SCAM-

sensitive residues, being somewhat more sensitive in the G protein over-expression

settings (Table 4.3 and Figure 4.).

80

Table 4.3 Changes in inhibition upon the coupling of G proteins Gα16 and Gαi2

EL2 Inhibition ± Inhibition TM7 Inhibition ± Inhibition Mutantsa number numberb Mutantsa number numberb KOR· KOR· KOR· KOR· KOR KOR Gα16 Gαi2 Gα16 Gαi2 C3157.38S 0.05 ± 0.03 0.04 0.13 C3157.38S 0.05 ± 0.03 0.04 0.13 C315S- C315S- -0.13 ± 0.05 0.17 0.43 -0.13 ± 0.09 0.37 0.28 V205C L3097.32C C315S- C315S- 0.06 ± 0.09 0.01 0.14 0.32 ± 0.06 0.14 0.13 D206C S3107.33C C315S- C315S- -0.11 ± 0.08 0.21 0.21 0.22 ± 0.06 0.17 0.24 V207C S3117.34C C315S- C315S- 0.24 ± 0.08 -0.17 0.07 -0.17 ± 0.03 0.10 0.26 I208C Y3127.35C C315S- C315S- 0.07 ± 0.02 0.01 0.26 0.28 ± 0.08 0.04 0.13 E209C Y3137.36C C315S- C315S- 0.05 ± 0.02 0.15 0.22 0.54 ± 0.04 0.05 0.18 S211C F3147.37C C315S- C315S- 0.57 ± 0.01 0.01 0.01 0.46 ± 0.04 0.10 0.21 L212C I3167.39C C315S- 0.003 ± C315S- 0.13 0.17 0.37 ± 0.03 0.02 0.19 Q213C 0.035 A3177.40C C315S- C315S- 0.39 ± 0.09 0.05 0.08 0.46 ± 0.05 0.03 0.25 F214C L3187.41C C315S- C315S- 0.22 ± 0.05 0.01 0.11 0.40 ± 0.06 -0.01 0.07 P215C G3197.42C C315S- C315S- 0.06 ± 0.04 0.05 0.14 0.65 ± 0.10 0 0.12 D216C Y3207.43C C315S- Average 0.06 0.17 -0.01 ± 0.02 0.18 0.23 T3217.44C C315S- 0.08 ± 0.04 0.13 0.34 N3227.45C TM6 Inhibition ± Inhibition C315S- 0.12 ± 0.04 0.07 0.31 Mutantsa number numberb S3237.46C KOR· KOR· C315S- KOR 7.47 -0.07 ± 0.04 0.16 0.35 Gα16 Gαi2 S324 C C315S- C3157.38S 0.05 ± 0.03 0.04 0.13 0.24 ± 0.04 -0.12 0.10 L3257.48C C315S- 0.46 ± C315S- 0.15 0.13 0.09 ± 0.04 0.26 0.30 E2976.58C 0.003 N3267.49C C315S- C315S- 0.28 ± 0.06 0.14 0.05 - - - V2966.57C P3277.50Cc

81

C315S- C315S- 0.67 ± 0.07 -0.16 -0.16 -0.06 ± 0.07 0.20 0.36 L2956.56C I3287.51C C315S- C315S- 0.65 ± 0.06 0.10 0.01 0.14 ± 0.01 0.13 0.23 I2946.55C L3297.52C C315S- C315S- 0.46 ± 0.06 0.03 0.18 0.14 ± 0.03 0.11 0.13 F2936.54C Y3307.53C C315S- C315S- 0.72 ± 0.06 -0.01 -0.03 0.11 ± 0.02 0.03 0.16 I2906.51C A3317.54C C315S- Average 0.04 0.03 0.18 ± 0.05 -0.02 -0.03 F3327.55C Average 0.09 0.20 aSCAM analysis revealed differential inhibition number changes in TM6, TM7 and EL2 of the KOR with various G protein coupling. MTSEA was used to react specifically with Cys side chains and the modified Cys show different inhibition ability of [3H]diprenorphine (~0.2nM) binding. The effects of MTSEA pretreatment on [3H]diprenorphine binding were expressed as inhibition number. Data shown represents the mean ±SEM of three to six experiments. Inhibition number was calculated according to Equation 1 in Chapter 2 MATERIALS AND METHODS. b“± Inhibition number” represents the difference of inhibition number between G protein- couplingand non-G coupling states. Negative sign (-) means decrease of inhibition number under G protein coupling; positive sign (omitted in Table 2) means increase of inhibition number. c[3H]diprenorphine binding was undetectable for the C3157.38S-P3277.50C mutant.

Figure 4.2a

KOR KOR⋅Gα 16 KOR⋅Gα i2 C315S L309C S310C S311C Y312C Y313C F314C I316C A317C L318C G319C Y320C T321C N322C hKOR TM7 S323C S324C L325C N326C P327C ** ** ** I328C L329C Y330C A331C F332C 0.00 0.25 0.50 0.75 0.00 0.25 0.50 0.75 0.00 0.25 0.50 0.75 [3H]Diprenorphine Binding Inhibition

Figure 4.2b

82

KOR KOR⋅Gα 16 KOR⋅Gα i2

C315S

E297C

V296C

L295C hKOR TM6 I294C

F293C

I290C

0.00 0.25 0.50 0.75 0.00 0.25 0.50 0.75 0.00 0.25 0.50 0.75 [3H]Diprenorphine Binding Inhibition

Figure 4.2c

KOR KOR⋅Gα 16 KOR⋅Gα i2

C315S

V205C

D206C

V207C

I208C

E209C

S211C hKOR EL2 L212C

Q213C

F214C

P215C

D216C

0.00 0.25 0.50 0.75 0.00 0.25 0.50 0.75 0.00 0.25 0.50 0.75 [3H]Diprenorphine Binding Inhibition

Figure 4.2 SCAM pattern. 2a SCAM analysis revealed differential MTSEA accessibility patterns in TM7 of the KOR upon various G protein-couplingconditions. MTSEA (1 mM for 5 min) was used to react specifically with Cys side chains and the modified Cys show different inhibition of [3H]diprenorphine (~0.2nM) binding. The effects of MTSEA pretreatment on [3H]diprenorphine binding were expressed as inhibition number. Each point represents the mean ±SEM of three to six experiments. Black bars indicate mutants for which inhibition numbers were significantly different (p < 0.01) from the reference (C3157.38S) by ANOVA and Dunnett’s test. 2b. SCAM analysis of the upper part of TM6. 2c. SCAM analysis of EL2. ** [3H]diprenorphine binding was undetectable for the C3157.38S-P3277.50C mutant.

83

4.2.2 Kinetics Studies Confirmed that Conformational Changes Occur

in KOR

The estimation of second-order rate constants of MTSEA reactions with sulfhydryls. was

performed by using the pseudo-first-order method (see Chapter 2 Materials and

Methods). Briefly, the extent of reaction after a fixed time with four concentrations of

MTSEA (all in excess over the reactive sulfhydryls) was determined. For kinetic studies, a slight fluctuation of reaction rate will affect the inhibition number dramatically due to the exponential relationship between the reaction rate and the inhibition effect. TM7 Cys mutants were chosen for kinetics study since they displayed larger average inhibition changes (0.2 under Gαi2 overexpression as discussed above). The reaction rate constants

varied significantly depending on the milieu ranging from 0.5- to 3.1-fold changes (Table

4.4) upon Gα16 overexpression. Most of the Cys mutants displayed no change in

reactivity or became more reactive, with two notable exceptions (N3227.45C and

7.52 7.49 L329 C). With the exception of N326 in the Gα16-overexpressed system, residue

positions closer to the extracellular side (i.e. top) of the KOR react more rapidly than do

those near the intracellular side when Gα proteins are overexpressed. For Gαi2 conditions, a different pattern of reactivity was observed. Some mutants showed increased reactivity (Y3137.36C, F3147.37C, L3187.41C and Y3207.43C) while L3297.52C

displayed decreased reactivity.

This kinetic method was not applied to the sensitive residues of TM6 and EL2. Generally

the second-order rate data were consistent with the SCAM data, showing a distinct

84 pattern of reaction rate constants for the sensitive mutants upon different G-protein coupling. We also examined the effect of pre-incubation with naloxone on protection against the MTSEA reagent. Our results were essentially similar to those previously published for both sensitive and insensitive residues (data not shown) (157, 158). For example, pretreatment with U69593 can block the inhibition induced by MTSEA in

7.49 N326 C, giving a protection number 31%, close to reported result ~38%(156). In Gαi2 overexpressed cells, for S3117.34C, Y3137.36C, N3227.45C, S3237.46C, and L3297.52C

mutants, no protection was observed using U69593.

Table 4.4 Second-order rate constants (k, M-1s-1) of MTSEA reaction with Cys mutants of KOR k, (M-1s-1) a ratiob ratioc

k (KOR·Gα16)/k k (KOR·Gαi2)/k KOR KOR·Gα16 KOR·Gαi2 (KOR) (KOR) C315S- 8.5 ± 1.1 7.8 ± 2.4 7.7 ± 0.9 0.9 0.9 S3107.33C C315S- 4.7 ± 1.0 7.1 ± 1.3 4.7 ± 0.3 1.5 1.0 S3117.34C C315S- 5.8 ± 1.7 7.1 ± 1.4 11.1 ± 3.2 1.2 1.9 Y3137.36C C315S- 4.2 ± 0.5 9.7 ± 2.4 6.8 ± 1.7 2.3 1.6 F3147.37C C315S- 6.1 ± 1.5 14.1 ± 3.1 7.9 ± 2.1 2.3 1.3 I3167.39C C315S- 4.7 ± 1.0 5.9 ± 1.3 4.2 ± 1.4 1.3 0.9 A3177.40C C315S- 4.5 ± 0.9 5.0 ± 1.1 8.1 ± 1.6 1.1 1.8 L3187.41C C315S- 5.8 ± 1.4 5.6 ± 1.1 4.7 ± 1.4 1.0 0.8 G3197.42C C315S- 5.2 ± 1.8 5.3 ± 1.7 8.6 ± 0.8 1.0 1.7 Y3207.43C C315S- 10.1 ± 5.3 ± 1.1 11.6 ± 3.2 0.5 1.2 N3227.45C 3.5 C315S- 4.7 ± 1.3 5.1 ± 0.8 6.1 ± 1.2 1.1 1.3 S3237.46C

85

C315S- 4.9 ± 0.8 15.2 ± 4.4 5.4 ± 2.0 3.1 1.1 N3267.49C C315S- 13.5 ± 7.3 ± 2.2 6.7 ± 1.5 0.5 0.5 L3297.52C 4.3 aSelected rates of Cys-mutant reaction with MTSEA were examined. Cells were treated with four concentrations of MTSEA, followed by quenching, washing [3H]diprenorphine binding. The second-order rate constants (k) were determined in triplicate. Data represent in mean ±SEM of three to eight independent experiments. b The ratio is k(KOR·Gα16)/k(KOR). c The ratio is k(KOR·Gαi2)/k(KOR).

4.2.3 Conformational changes induced by G protein-coupling have

significant effects on agonist affinities.

To examine the consequences of these conformational changes induced by various G

protein backgrounds, three KOR agonists (salvinorin A, U69593 and dynorphin A (1-13))

were tested against wild-type KOR transiently expressed in the three cell systems (Figure

1). Two different radioactive ligands (the antagonist [3H]diprenorphine and the agonist

3 [ H]U69593) were also used. In both the Gα16 and Gαi2 environments, salvinorin A and

U69593 demonstrated significantly enhanced affinities (Table 4.5) although, surprisingly,

the endogenous peptide dynorphin A (1-13)’s affinity was almost unchanged. Using

[3H]U69593 revealed a differential preference between salvinorin A and U69593 as well.

The largest effect (18-fold) was seen for salvinorin A in the Gα16 background using

[3H]U69593 as radioligand.

Table 4.5 Differential effects of G protein overexpression on agonists (salvinorin A,

U69593 and dynorphin A (1-13)) binding affinities (Ki, nM) Radioactive label [3H]diprenorphine Tested Agonistsc salvinorin A U69593 Dynorphin A (1-13) a b a b a b Ki (nM) ratio Ki (nM) ratio Ki (nM) ratio KOR 33 ± 8 69 ± 12 0.94 ± 0.21 KOR·Gα16 13 ± 4* 2.5 59 ± 8 1.2 0.98 ± 0.56 1.0

86

KOR·Gαi2 8.6 ± 1.5* 3.8 25 ± 6* 2.8 1.5 ± 0.7 0.6 Radioactive label [3H]U69593 Tested Agonistsc salvinorin A U69593 Dynorphin A (1-13) a b a b a b Ki (nM) ratio Ki (nM) ratio Ki (nM) ratio KOR 0.80 ± 0.33 1.0 ± 0.2 0.19 ± 0.06 KOR·Gα16 0.045 ± 0.019* 18 0.81 ± 0.13 1.2 0.22 ± 0.07 0.9 KOR·Gαi2 0.12 ± 0.06* 6.7 0.89 ± 0.15 1.1 0.21 ± 0.02 0.8 a The affinity constants (Ki, nM) of the different agonists were determined in competition binding assays with [3H]diprenorphine (antagonist) or [3H]U69593 (agonist) and increasing concentrations of agonists (from 10-5 nM to 104 nM). Data represent three to six independent experiments in mean ±SEM. b The ratio is Ki(KOR)/Ki(KOR·Gα16) or Ki(KOR)/Ki(KOR·Gαi2). cSalvinorin A and U69593 are small-molecule agonists (MW ~400 Da); dynorphin A (1-13) is the short form of the endogenous peptide agonist (MW ~1600 Da). *p < 0.05 vs KOR

4.2.4 Refining the Binding Site of Salvinorin A

It has been previously suggested that EL2 is critical for salvinorin A-KOR binding although no residues have been thus far identified. As the sensitive residues identified by

SCAM, L212 and F214 have great potential to be critical residues for salvinorin A binding, mutagenesis was applied to them. Because they are quite bulky hydrophobic residues, L212A and F214A, were constructed to examine a potential steric hindrance effect. Interestingly, substitution at L212 and F214 enhanced salvinorin A affinity

Ki values (1.7 nM and 1.0 nM respectively; Table 4.6) while alanine mutagenesis of neighboring residues (S211A and Q214A) had no effect. According to our recent model, the EL2 dips down into the hKOR binding pocket (Figure 4.3), placing L212 and F214 into the binding pocket. Some other critical residues (I2946.55A and Q1152.60A) were also

2.60 examined. Q115 A decreased the Ki value 17 fold, consistent with the results reported by Kane et al.(212). The role of Q1152.60 in salvinorin A binding is not well understood; however, a hydrogen bond with the furan oxygen of salvinorin A is suggested by our latest models (Figures 4.3a and 4.3b).

87

Table 4.6 Affinity (Ki, nM) of salvinorin A binding to WT KOR and mutants c Kd Bmax Ki (sal A) ratio a a b (nM) (pmol/mg) (nM) Ki(Mutant)/Ki(WT) KOR-WT 0.46 ± 0.10 2.0 ± 0.4 33 ± 8 S211A 0.47 ± 0.10 2.7 ± 0.8 38 ± 5 1.2 L212A 0.28 ± 0.07 0.27 ± 0.08 1.7 ± 1.0 0.05 Q213A 0.51 ± 0.15 0.76 ± 0.04 40 ± 12 1.2 F214A 0.46 ± 0.22 0.8 ± 0.7 1.0 ± 0.1 0.03 I2946.55A 0.53 ± 0.18 0.9 ± 0.8 12 ± 6 0.4 E2976.58A 0.91 ± 0.08 1.8 ± 1.2 15 ± 3 0.5 Q1152.60A 0.65 ± 0.10 2.5 ± 1.7 556 ± 231 17 aSaturation binding of [3H]diprenorphine to the wild type and the mutants was performed according to the procedure in MATERIALS AND METHODS. Data represent in mean ±SEM of two to four independent experiments. b The affinity constants (Ki) of the different compounds were determined in competition binding assays with [3H]diprenorphine and increasing concentrations of salvinorin A (from 10-5 nM to 104 nM). Data represent the mean ± SEM of two to four independent experiments. c The ratio is Ki(Mutant)/Ki(WT).

After considering the SCAM data and the recently determined crystal structure of

photoactivated bovine rhodopsin (23), an updated model for the activated form of the

hKOR, as recognized by the agonist salvinorin A, is shown in Figures 4.3. The sidechain

conformations in the TM helical regions of the ‘inactive state’ (PDB 1U19(183)) and

‘activated’ rhodopsin crystals are essentially the same; however, there is a subtle but distinct difference in the positions of the TM helices relative to one another. In the

‘activated’ crystal form, TM3 and TM6 are roughly 1 Å further apart from one another in

the intracellular region as compared to the ‘dark state’. Based upon the experimental

findings reported here and elsewhere(185, 187), the binding of the G protein complex is

thought to induce an active or ‘active-like’ state that is more conducive to the binding of

agonists than is the inactive state. This change upon activation (or binding of G protein)

of the GPCR effectively extends the binding pocket to include a narrow region bounded

by the intracellular regions of TM2, TM3, TM6 and TM7. As shown in Figure 4.3b, the

residues of TM7 that line this narrow extended binding site are those residues that, under

88 conditions of G protein overexpression, were found to significantly affect the binding of

[3H]diprenorphine when those positions were subjected to the SCAM procedure. This implies that these residues become accessible to the relatively small MTSEA reagent when G proteins are overexpressed, and this is reflected in the ‘activated’ rhodopsin- based model.

3a 3b

Figure 4.3 KOR model based on active rhodopsin structure. 3a. Ribbon diagram of an earlier hKOR model derived from the ‘dark’ rhodopsin crystal structure. 3b. Ribbon diagram of the activated hKOR model derived from the ‘light’ rhodopsin crystal structure. Ribbons are color-coded based on secondary structure: red = α helix, blue = β strand, violet = turn, yellow = coil. Selected sidechains of residues in TM6, TM7 and EL2 whose positions were tested for [3H]diprenorphine inhibition are shown as capped sticks. Sidechain color coding indicates the conditions under which significantly altered [3H]diprenorphine binding occurred relative to the C3157.38S reference (see Figures 4.2a- . . . . c): green = KOR, KOR Gα16 and KOR Gαi2; yellow = KOR Gα16 and KOR Gαi2; red = . . KOR Gαi2 only; magenta = KOR and KOR Gα16. Sidechains containing grey atoms did not significantly alter [3H]diprenorphine binding under any of the three conditions or were not tested (i.e. C3157.38). Cyan indicates conserved residues. A Connolly channel plot delineating the binding pocket in each receptor is shown as a transparent grey surface (probe radius = 1.4 Å). TM6 and TM7 are closest to the viewer.

89

In summary, significant conformational changes were observed on TM7, the extracellular portion of TM6 and EL2. Eight SCAM-sensitive residues (S3107.33, F3147.37 to Y3207.43) on TM7 presented a cluster pattern when the KOR was exposed to baseline amounts of G protein, and additional residues became sensitive upon over-expression of various G

7.34 7.49 proteins. In TM7, S311 and N326 were found to be sensitive in Gα16-overexpressed

7.36 7.45 7.46 7.52 cells and Y313 , N322 , S323 and L329 in Gαi2-overexpressed cells. In

addition, the degree of sensitivity for various TM7 residues was augmented, especially in

Gαi2-overexpressed cells. A similar phenomenon was also observed for residues in TM6

and EL2. In addition to enhanced sensitivity of certain residues, our findings also

indicated that a slight rotation was predicted to occur in the upper part of TM7 upon G

protein overexpression. These relatively modest conformational changes engendered by

G protein overexpression had both profound and differential effects on the abilities of

agonists to bind to KOR. These data are significant because they demonstrate that Gα

subunits differentially modulate the conformation and agonist affinity of a prototypical

GPCR.

4.3 Discussion

In the way it is applied here, SCAM is a biochemical method capable of detecting averaged receptor conformations in a defined time period (5 min in these studies). Both

static conformations as well as averaged conformational fluctuations during the period

the system is exposed to the MTSEA reagent will be reflected in altered Cys reactivity.

The final SCAM pattern of MTSEA sensitivity thus represents a summary of the

multiplicity of conformations of KOR in the presence and absence of Gα subunit

90 overexpression. Our data in Table 4.3 demonstrate that the averaged inhibition values of

SCAM-sensitive residues are significantly changed by Gα subunit overexpression— especially by Gαi2. This change in MTSEA sensitivity in the absence of agonist binding

strongly supports the notion that Gα subunits can induce conformational changes in

GPCRs. It has been conventionally postulated that following agonist exposure GPCR

conformations are altered. Indeed, some novel methodological approaches have recently

demonstrated agonist-induced conformational changes in several GPCRs (213-215). To

our knowledge, our results are the first to demonstrate that G proteins can alter GPCR

conformations in the absence of ligand and, in addition, these findings are consistent with

a growing literature suggesting the existence of precoupled GPCR-G protein complexes

(187, 216, 217).

In this study, the patterns of residue accessibility (see Figures 4.2a-c) were rationalized

via the comparison of our hKOR model with inactive and active rhodopsin structures.

Residues L3097.32 to Y3137.36 are located at the extracellular end of TM7. L3097.32 and

Y3127.35 were never accessible (or undetectable by inhibition of binding) while S3107.33 was always accessible and significantly affected the binding of diprenorphine. S3117.34

7.36 became accessible upon binding of Gα16 or Gαi2. Y313 became accessible only on

overexpression of Gαi2. In the regions comprising the extracellular portions of both TM6

and TM7, there are contiguous sequences of amino acids (F2936.54 to L2996.60 and

F3147.37 to Y3207.43) that were accessible. This cluster pattern of accessibility has

previously been observed in TM6 and TM7 for the MOR and DOR, as well as the KOR

(157, 158). However, some of the most sensitive residues are not facing directly into the

91 binding pocket (Figure 4.2a). Further study is necessary to clarify this. To date, this pattern of accessibility has been generally attributed to the inherent flexibility of the helix and/or to the permeability of the MTSEA reagent through the lipid membrane (218). The over accessibility patterns are likely due to the disorder of this region of the helix, also dynamic movements and the proline kink at P3277.50 may play a role. In a recent

study(157), Xu et al. also suggested that the large distance between TM1 and TM7 may

allow MTSEA to penetrate the KOR TM1-TM7 interface to react with substituted Cyss.

The most sensitive residues on TM6 (E2976.58, I2946.55 and I2906.51) and EL2 (L212 and

F214) are pointing to the binding pocket, which is consistent with the current model.

A pseudo-first-order method was used for obtaining the second-order rate of MTSEA

reactions. For the 13 TM7 Cys mutants examined (Table 4.4), most (S3117.34C,

Y3137.36C, F3147.37C, I3167.39C, L3187.41C, Y3207.43C and N3267.49C) have increased or

unchanged reaction rate constants under either Gαi2 or Gα16 overexpression systems,

which are consistent with the SCAM data. S3117.34C, I3167.39C and N3267.49C show enhancement under Gα16 overexpression but not Gαi2. The differences may be caused by

an unfavorable local environment change presented in the new conformations in the Gαi2 overexpression system, but could also be caused by the proposed inherent flexibility of the extracellular portion of TM7. Since the conformational change is not dramatic and those residues are located in the upper part of the helix, a longer time exposure of

MTSEA could compensate and reveal a high SCAM inhibition. N3227.45C, S3237.46C and

L3297.52C, whose access to MTSEA is presumably less than that of more extracellularly-

located residues (even under conditions of G protein overexpression), and whose kinetic

92 reaction rates are generally less than those of more extracellularly-located residues, can still significantly affect the binding of diprenorphine when these mutants are reacted with

MTSEA. L3297.52C is furthest down in the pocket and theoretically least accessible to

MTSEA (see Figure 4.3b). It should be mentioned that the mechanism by which the

MTSEA reaches buried Cys residues such as C3297.52 is not perfectly clear, as some

studies have indicated that MTSEA is lipid membrane-permeable (218). In general, the

reaction rates with G protein overexpression are within 0.5- to 3.1-fold of the

corresponding non-G protein overexpression systems, and there is no large increase of

rate observed between different mutants. Analysis of SCAM second-order reaction rate

constants had been used to predict which residues most likely (or most often) face the

interior of the binding cavity (158). Some of the reaction rate constants reported here

were lower than previously reported (157, 158) and are likely due to differences in

experimental techniques. In addition, the reactivity of individual residues (in TM7) did

appear to indicate a directional preference and α-helical periodicity (see Figure 4.4).

Figure 4.4 shows that TM7 may undergo a modest counterclockwise rotation (viewed

from the extracellular side) when in the presence of overexpressed Gαi2, but the rotational preference for TM7 of the Gα16-overexpressed KOR is less clear; its pattern of

reactivity is possibly due to the effects of MTSEA membrane permeability and/or

rotational flexibility. However, it seems reasonable to expect that the KOR presents a

consistent and complementary receptor conformation to approaching agonists, at least in

the case of Salvinorin A and U69593, whose binding affinities increased by roughly the

same amount in both Gα16 and Gαi2 overexpressed systems (see Table 4.5).

93

Side View: KOR•Gα16 KOR•Gαi2

Top View: KOR•Gα16 KOR•Gαi2

Figure 4.4 Second-order rate constants (k, M-1s-1) revealed a directional preference and α-helical periodicity. The ratios (k(Gα16)/k(non-Gα) or k(Gαi2)/k(non-Gα)) of MTSEA reaction rates upon G protein-coupling were divided into three subgroups, highlighted by labeling the alpha carbon with different colors: Red: ratio > 1.5; green, 0.5 < ratio <= 1.5; blue, ratio <= 0.5. All of the residues presented inhibited the binding of [3H]diprenorphine by greater than 35% upon Gαi2 coupling, and were significantly different from the

94 background C3157.38S in SCAM pattern (see Figure 4.2a). The KOR backbone is represented by a thin ribbon in which the individual TMs are color-coded (TM1 = red; TM2 = orange; TM3 = yellow; TM4 = green; TM5 = cyan; TM6 = blue; TM7/Helix 8 = violet), and salvinorin A is rendered as capped sticks (carbon = grey; oxygen = red).

Compared to TM6 and TM7, the pattern of EL2 is relatively simple, with only two

residues (L212 and F214) next to C210 — which likely forms a disulfide bond with C131

— identified as sensitive. The residues next to C210, L212 and F214 are likely not to be as conformationally flexible as the other loop residues, and are more likely to present a fixed direction pointing down into the binding pocket. Similar observations were reported for the D2- EL2 (219), where two residues (I184 and N186) were

found to be SCAM-sensitive. In the D2 receptor, I184 and N186 are adjacent to the

disulfide bond-forming C182, just as L212 and F214 are adjacent to C210 in the KOR.

The KOR modeling studies presented here led to the prediction that L212 and F214

would cause a large steric hindrance for ligand binding to KOR. A reasonable prediction

was that a smaller hydrophobic residue will lessen the blockage and potentiate ligand

binding. The profound increase of salvinorin A binding for KOR mutants (L212A and

F214A) supports this model (Table 4.6).

In summary, our major finding is that Gαi2 and Gα16 differentially modulate the

conformation of a prototypical family A GPCR — KOR. It was discovered that when

Gαi2 and Gα16 were over-expressed together with KOR, a differential pattern of Cys-

accessibility to MTSEA was observed. In conjunction with these apparent changes in

KOR conformation there was a selective alteration in salvinorin A’s affinity. These

results are consistent with the hypothesis that Gα subunits differentially interact with

95

GPCRs in the absence of agonists and that such interactions lead to conformational changes which are reflected in altered ligand affinities. Together with a recent chimeric- receptor/SCAM study by Vrotherms et. al.(160), showing TM2’s orientation varies among opioid receptor family members, these findings revealed the mechanism why salvinorin A binds to KOR with such high affinity as well as high selectivity: the availability of both the critical residues (Y313, Y320, Y119 and Q115 etc.) and the appropriate helical orientation and conformation of the TM domains where these critical residues are located in (such as TM7 and TM2).

96

Chapter 5: The Design and Application of Covalently-Bound Agonists to Probe KOR

5.1 Introduction and Rationale

Salvinorin A, the most potent natural hallucinogen, gained immediate interest since KOR was identified as the sole molecular target in 2002 (1). Salvinorin A represents an attractive lead compound for as discussed in Chapter 3. During the past few years, more than 250 derivatives of salvinorin A were synthesized by exploring almost all the commonly seen functional groups at C-2, C-4 and C-12 positions (Figure

3.1). Some of these analogs presented unique pharmacological profiles, from full KOR agonist to partial DOR or MOR agonists and antagonists (5, 220-224). A recent paper showed that the modification of C-12 of the furan ring yielded analogs with promising

KOR antagonistic activity (225). However, most of the 250 analogs showed moderate affinity or even completely lost affinity to KOR. The challenge now is how to gain precise knowledge about salvinorin A – KOR interactions and rationally design novel salvinorin A derivatives. Covalent binding, also referred to as affinity labeling or proximity-accelerated chemical coupling, emerged as a powerful approach to unambiguously characterize ligand’s binding site in recent years. This method takes advantage of the chemically reactive amino acids inside or close to the ligand binding site, most of the time the reactive residue is a nucleophile (e.g., Cys). Introduction of a

Cys-preferred reactive group (i.e., an electrophile) into the ligand structure leads to a covalent reaction between the ligand and receptor if the chemical reaction conditions are favorable (Figure 5.1). The reactive Cys residues could already exist in the protein or be

97 introduced by site-directed mutagenesis, which provides the opportunity to study receptors that lack Cys residues inside their binding sites. Because of the specific recognition between the ligand and receptor, this affinity labeling is also expected to be highly selective. By adjusting the reactive functional group on the covalently-bound ligand, a wide range of labeling reaction rates and structurally diverse labeling products can be obtained. Halogen and isothiocyanate groups are commonly used for such covalently-bound ligand design due to their relatively high reactivity under mild physiological conditions. The successful applications of affinity labeling methods include an (imidazobenzodiazepine) for GABAA receptor (226),

fluorescently tagged inhibitors for calcium-bound protein deiminase 4 (PAD4)

(227), selective kinase inhibitors (228, 229), estradiols for receptor (230-235),

and antagonists for N-methyl-D-aspartate (NMDA) receptor (a ligand-gated cation

channel) (236).

To design covalently-bound ligands, we had to find a chemically reactive amino acid

inside the binding site of salvinorin A. Our earlier binding site study showed there was a

direct interaction between the acetoxy group of salvinorin A and Y3137.36, a critical residue inside the binding pocket (2). Nearby, a water-accessible C3157.38 is highly reactive to methanethiosulfonate reagent, which has been shown in the SCAM study by

Xu et al. (176). Because the C3157.38S mutation only affects the binding affinity

moderately, it is very possible that C3157.38 exists as a free Cys rather than as a

contributor to global structure or disulfide-bonds. Therefore this free and chemically

reactive C3157.38 is a potential nucleophile at physiological pH 7.4 and can provide an

98 anchoring point for affinity labeling (Figure 5.1). Our hypothesis is that salvinorin A analogs with an appropriate electrophilic group will possibly react with C3157.38 if these two reactive groups are within close proximity. The product of affinity labeling, i.e., the modified KOR protein, can be characterized by both the traditional pharmacological and novel mass spectrometric techniques (230, 231).

Figure 5.1 The mechanism for RB-64 labeling (covalently-bound to) KOR. The molecular weight shift after the substitution reaction is about 431.5.

To test our hypothesis, we designed a research plan involving several technically demanding steps, including: 1) design and synthesize chemically reactive salvinorin A derivatives which can label a Cys residue under physiological conditions, i.e., pH 7.4 and mild temperature; 2) express a large quantity of KOR protein in a purification-friendly construct (e.g., FLAG and His6 tagged); 3) optimize the labeling reaction between

covalently-bound ligand and KOR, and apply the optimal conditions to a large quantity of

KOR (e.g., 20-150 mm dish cells transiently expressing KOR, labeling with RB-64 for 3

hr/10µM at 4°C in PBS buffer); 4) isolate the RB-64 modified KOR protein by tandem

2+ affinity purification (i.e., M2 resin for FLAG tag, Ni -NTA resin for His6 tag); 5)

concentrate the modified KOR protein and confirm the existence of KOR by western

99 blotting (i.e., anti-FLAG and anti-KOR antibodies), then use SDS-PAGE gel and

Coomassie staining to separate and detect KOR from other associated proteins; 6) use an appropriate protease to perform in-gel digestion (i.e. Chymotrypsin to cover the TM7 region) and identify the labeling site (C3157.38) through mass spectrometry; 7) validate the affinity labeling site(s) by mutagenesis studies of the residues spatially close to

C3157.38, such as Y3137.36, F3147.37 and I3167.39.

Thus far, two compounds — RB-48 and RB-64 (both with pM affinities and extraordinary selectivity for KOR) — have emerged as suitable ligands for affinity labeling. RB-64, RB-48 and related compounds were synthesized by our collaborator Dr.

Jordan K. Zjawiony (University of Mississipi). The structures and possible labeling mechanism of RB-64 is shown in Figure 5.1. Overall, this study validated our prior salvinorin A binding site in KOR (2), and provided even more precise information about the ligand-receptor interactions. Undoubtedly this study will further aid the rational design of chemically unique KOR ligands with therapeutic potential. Moreover, this study predicts a bright future for the application of mass spectrometry in GPCR research.

5.2 Results

5.2.1 Characterizing Covalently-Bound Ligands by Molecular Pharmacology Methods 5.2.1.1 RB-64 and RB-48 are potent KOR agonists

The binding affinities of RB-64 and RB-48 were defined by competition binding assays with both [3H]U69593 (selective agonist for KOR) and [3H]diprenorphine (non-selective

100 antagonist for KOR). RB-64 and RB-48 showed affinities similar to salvinorin A in competing with [3H]diprenorphine. However, [3H]U69593 can differentiate RB-64 and

RB-48: the latter showed a similar affinity (2.10 nM) as salvinorin A’s (1.85 nM); while

RB-64 showed a moderate 3-fold increase (0.59 nM) in binding affinity as compared to

salvinorin A. Both compounds presented high potency in the [35S]GTPγS assay.

Interestingly RB-64 had an EC50 value down to 0.077 nM, much smaller than salvinorin

A’s (17 nM) and RB-48’s EC50 was in between them (0.19 nM).

Table 5.1 The pharmacological profile of RB-64 and RB-48

a b c Ki (nM) Ki (nM) EC50 (nM) SalA 1.85± 1.49 21 ± 11 17 ± 6 RB-48 2.10 ± 0.84 32 ± 15 0.19 ± 0.01 RB-64 0.59 ± 0.21 39 ± 11 0.077 ± 0.016

a The affinity constants (Ki) of the different compounds were determined in competition binding assays with [3H]U69593 and increasing concentrations of unlabeled compounds. Each value is the mean of three independent experiments. b The affinity constants (Ki) of the different compounds were determined in competition binding assays 3 with [ H]diprenorphine. c 35 The functional potency (EC50) of the different compounds were determined in [ S]GTPγS assay. See Chapter 2 for details.

5.2.1.2 Covalent labeling of WT KOR and mutants with RB-64

Our earlier covalent labeling study of RB-64 and RB-48 showed that both ligands can

covalently bind to KOR. The labeling result of RB-48 was not consistent due to its high

reactivity and low stability. RB-64 was less reactive, but consistently demonstrated a

higher labeling ability. A time course and dose response study of RB-64 labeling showed

that 3hr/10µM (at 4°C in PBS buffer) was the optimal condition for effective RB-64

101 labeling without the non-specific effects caused by both higher concentrations and longer incubation times. At these conditions, a maximal 59 % of KOR was labeled by RB-64, and 41% residual binding left (Figure 5.2a). As comparison, both salvinorin A and naloxone didn’t show any significant labeling of KOR in parallel experiments (Figure

5.2a). However, there was a slight drop in Bmax value for salvinorin A labeling, which was

probably caused by inefficient washing (intact cells used) and non-specific hydrophobic

interactions. Stronger reaction conditions, i.e., 10hr/20µM, achieved 82% labeling of

KOR. To avoid non-specific effects associated with high dose of RB-64, we used the

moderate labeling conditions (i.e., 3hr/10µM) in the irreversible binding study.

To precisely determine the labeling site of RB-64, WT KOR with native C3157.38 and

mutated KOR with mutations C315S, Y313C, F314C and I316C were tested using the

RB-64 labeling assays. Interestingly, RB-64 preferred residues close to Y3137.36 —

7.37 7.38 7.39 F314 and C315 — but not I316 , which was too far away. The residual KOR Bmax values for Y3137.36 and I3167.39 (> 200% after normalization) are much larger than their

corresponding internal standards (100% refers to untreated KOR). And intriguingly, the

labeling silent mutation C315S had no significant reactivity with RB-64 as compared to

naloxone labeling, even though all three compounds (RB-64, salvinorin A and naloxone)

decreased the Bmax of KOR moderately. The mechanism for this unexpected effect is not

clear. Insufficient washing could be one of the reasons. By comparing the residual Bmax values, the reactivity of Cys mutants to RB-64 was lost for I316C, Y313C and C315S mutations. All these data are consistent with our prior model, in which C-2 position of

102 salvinorin A is within close contact to Y3137.38, so residues C3157.36 and F3147.37 are in

the most favorable position for labeling reaction.

6000

5000

4000

3000 fmol/mg max,

B 2000 * 1000 * 0 e e -64 -48 B B R Sal A RB-64 R Sal A NaloxonNO Label NaloxonNO Label 10μM, 3hr 20μM, 10hr

2a

RB-64 Sal A # Naloxone 350

300 # 250

% 200 max B 150

100 # 50 # 0 C315S Y313C F314C C315(WT) I316C

2b Figure 5.2 RB-64 labeling. Irreversible reaction of RB-64 with (2a) WT KOR and (2b) Cys mutants (based on C315S background). (2a) Receptors were exposed for 3h/10µM

103

RB-64, sal A, naloxone and no drug treatment (vehicle DMSO only). (2b) 10 hr/20 µM RB-64, sal A, Naloxone and Blank (vehicle only) were used for labeling. After labeling reaction, the cell membranes were extensively washed (three times); the residual binding 3 was determined by [ H]diprenorphine saturation binding. In (2b) Bmax values were expressed as a percentage of the internal control in which no drug, only vehicle was present in the PBS buffer. Each bar represents mean ± SEM for two to four independent experiments in which each sample was measured in triplicate. See Chapter 2 for details.

5.2.2 Characterizing Covalently-Bound Ligands by Mass Spectrometry

5.2.2.1 KOR expression, labeling and purification

To improve the efficiency of purification, a tandem tagged KOR, containing an N- terminal FLAG tag and C-terminal His6 tag, was constructed in pcDNA3.1(+) vector and

expressed in HEK293 T cells. Cells were harvested after being labeled by RB-64 for 10

hr/20µM. Cell lysates were purified by sequential Anti FLAG M2 and Ni2+-NTA affinity

purification. The elute was concentrated via Millipore filtration tubes with MW 10 kDa

cutoff. The presence of KOR proteins was confirmed by western blotting and Coomassie

staining after electrophoretic separation (Figure 5.3). The final products in western

blotting showed two KOR bands: 1) “mature type” KOR at 52 kDa and 2) “immature

type” KOR at 47 kDa (Figure 5.3a). The distribution of KOR protein in two MW bands

was also observed by Wannemacher et. al (237). The coomassie stained gel was then submitted to the UNC-Duke Michael Hooker Proteomics Center at UNC, Chapel Hill for mass spectrometric analysis. The correct protein bands (i.e., 52 and 47 kDa) were both excised from the gel, digested by chymotrypsin and examined by mass analyzers. Two mass spectrometers, AB 4800 MALDI TOF/TOF and Bruker Ultraflex MALDI

TOF/TOF, were used for our study. Both mass analyzers can detect femtomole quantities

of peptides and proteins.

104

52 kDa 52 kDa 47kDa

47kDa

Anti-FLAG Anti-KOR

3a 3b Figure 5.3 Separation of KOR by SDS-PAGE. (3a) western blotting (10% SDS-PAGE) of purified FLAG-KOR-His-tagged receptor after two-step purification by anti-FLAG M2 and Ni2+ NTA resins. Double bands (52 kDa and 47 kDa) represent various degree of glycosylation of KOR protein (237). (3b) Coomassie stained 4-20% SDS-PAGE gel.

5.2.2.2 Mass spectrometric analysis

Peptide samples, containing KOR fragments after chymotrypsin digestion, were analyzed

on automated AB 4800 MALDI TOF/TOF and Bruker Ultraflex MALDI TOF/TOF. Both instruments gave similar patterns of identified chymotryptic peptides. For each MALDI analyzer, a small amount of sample was co-crystallized with α-cyano-4-hydroxycinnamic acid (CHCA) and was analyzed over an m/z range of 600-4000. In Figure 5.5a, the spectrum of chymotrypsin is shown to define those peptides that represent autolysis products found in the experimental samples. The mass spectra generated from chymotryptic digestion of the 52 and 47 kDa KOR bands are shown in Figures 5.5b and

5.5c. The Mascot searches of these chymotryptic peptides identified human KOR

105

(accession number AAA63646, JC2338, Q8IWP3 human, Q499G4 human) as the only protein hit with significant scores found in both the higher and lower MW bands (Table

5.2). 19 matched peptides were found for high score JC2338 (Table 5.3), various oxidation states of the same peptides were omitted. The sequence coverage of KOR was

26%, including the potential anchoring site for RB-64 – C3157.38 (Figure 5.5 and Table

5.3).

Table 5.2 Mascot searches of chymotryptic peptides from KOR (47 kDa) and KOR (52 kDa). Protein scores greater than 64 are significant (p < 0.05). Accession Mass Score Description

1. JC2338 42631 77 kappa opioid receptor 1 - human

KOR DRG kappa 1 splice variant KOR 1A.-

47 2. Q8IWP3_HUMAN 41087 71 Homo sapiens (Human). kDa

Opioid receptor, kappa 1.- Homo

3. Q499G4_HUMAN 42617 70 sapiens (Human).

1. AAA63646 16607 77 HUMKOR NID: - Homo sapiens

DRG kappa 1 splice variant KOR 1A.-

2. Q8IWP3_HUMAN 41087 77 KOR Homo sapiens (Human). 52 kDa 3. JC2338 42631 75 kappa opioid receptor 1 - human

Opioid receptor, kappa 1.- Homo

4. Q499G4_HUMAN 42617 75 sapiens (Human).

106

Table 5.3 Detected KOR peptides from chymotryptic digestion by AB 4800 MALDI TOF/TOF Peptide Observed Expected Missed position monoisotopic monoisotopic Sequence Ions Cleavage Start-End [MH]+, Da [M]+, Da 1 - 8 964.56 963.47 MESPIQIF 0 67 - 74 819.49 818.49 SVVFVVGL 1 71 - 79 857.48 856.50 VVGLVGNSL 1 75 - 82 882.46 881.43 VGNSLVMF 1 148 - 157 1248.58 1247.55 TLTMMSVDRY 1 150 - 157 1034.47 1033.42 TMMSVDRY 0 168 - 173 748.43 747.39 DFRTPL 30 1 226 - 233 1000.59 999.53 MKICVFIF 1 232 - 240 1018.53 1017.63 IFAFVIPVL 2 247 - 253 875.53 874.53 TLMILRL 2 260 - 269 1161.61 1160.59 SGSREKDRNL 16 0 300 - 309 931.53 930.44 GSTSHSTAAL 0 314 - 320 786.44 785.38 FCIALGY 1 370 - 380 1257.64 1256.65 LRDIDGMNKPV 1 371 - 380 1160.56 1159.57 RDIDGMNKPV 0

MKTIIALSYIFCLVFA DYKDDDDK MESPIQIFRGEPGPTCAPSACLPPNSSAWF Signal sequence FLAG 10 20 30

PGWAEPDSNGSAGSEDAQLEPAHISPAIPVIITAVYSVVFVVGLVGNSLV 40 50 60 70 80

MFVIIRYTKMKTATNIYIFNLALADALVTTTMPFQSTVYLMNSWPFGDVL 90 100 110 120 130

CKIVISIDYYNMFTSIFTLTMMSVDRYIAVCHPVKALDFRTPLKAKIINI 140 150 160 170 180

CIWLLSSSVGISAIVLGGTKVREDVDVIECSLQFPDDDYSWWDLFMKICV 190 200 210 220 230

FIFAFVIPVLIIIVCYTLMILRLKSVRLLSGSREKDRNLRRITRLVLVVV 240 250 260 270 280

AVFVVCWTPIHIFILVEALGSTSHSTAALSSYYFCIALGYTNSSLNPILY 290 300 310 320 330

AFLDENFKRCFRDFCFPLKMRMERQSTSRVRNTVQDPAYLRDIDGMNKPV 340 350 360 370 380

HHHHHH His6

107

Figure 5.4 Sequence coverage of KOR by MALDI TOF/TOF. FLAG and His6 tags are indicated with rectangular boxes. The solid black lines represent the chymotrypsin digested peptides detected by MS analysis. The thicker solid black lines indicate that the corresponding sequences were confirmed in MS/MS analysis by AB 4800 MALDI TOF/TOF. Cys315 is shown in bold and red color. All TM domains are highlighted by the grey background. A total 26% sequence coverage was observed for KOR 47 kDa, 19% sequence coverage for KOR 52 kDa.

Theoretically RB-64 modification can add about 431.5 Da to the peptide fragment containing C3157.38. Because the modification by RB-64 was not stocked in the database,

the modified peptide was likely found in the pool of unmatched peptides after the Mascot

searches. One unmatched peptide, m/z 1381.80, was close to the modified peptides

predicted by in silico digestion as shown in Table 5.4. An ongoing MS/MS analysis of

this potential modified peptide (YFCIALGY) preliminarily confirmed the covalent

modification of C3157.38 (data not shown).

108

Chymotrypsin

5a KOR (52 kDa)

5b

109

KOR (47 kDa)

5c

Figure 5.5 MALDI mass spectra of KOR after chymotrypsine digestion. (5a) chymotrypsin (5b) KOR 52 kDa (5c) KOR 47 kDa are analyzed by Bruker Ultraflex MALDI TOF/TOF. In (5b) and (5c), x-axis zoom inserts were showing RB-64 modified YFCIAGY (MW 1381.8) as compared to unmodified (MW 949.5) peptides. The strong peak (1523.7) indicates the chymotrypsin autolysis.

Table 5.4 RB-64 modified peptides predicted by in silico chymotrypsin digestion

Peptide Unmodified Modified Missed position monoisotopic monoisotopic Sequence Cleavage Start-End [MH]+, Da [MH]+, Da

315 - 320 639.4 1070.9 CIALGY 0

314 - 320 786.4 1217.9 FCIALGY 1

313 - 320 949.5 1381.0 YFCIALGY 2

315 - 330 1741.9 2173.4 CIALGYTNSSLNPILY 1

110

314 - 330 1889.0 2320.5 FCIALGYTNSSLNPILY 2

CIALGYTNSSLNPILY 315 - 332 1960.0 2391.5 2 AF

5.2.3 in vivo PPI Study of RB-64

Prepulse inhibition (PPI) means the inhibition of a startle reflex when a prepulse happens

before the startling stimulus, such as a loud sound (238-241). Clinical studies have shown

that schizophrenic patients have deficient PPI. PPI can be disrupted by psychomimetics, and the reversal of this impaired PPI was used to predict potential drugs.

Because the parent compound of RB-64 is salvinorin A, the most potent naturally

occurring hallucinogen (1), we predicted that both RB-64 and salvinorin A have effects

on PPI. And we observed an extraordinarily high potency of RB-64 as compared to salvinorin A. Experimental details are shown in chapter 2. This part of work was performed in collaboration with Dr. William Wetsel (Duke University).

Null Activity — Salvinorin A and RB-64 produced changes in activity during null trails

(without any stimuli) compared to vehicle-treated mice (Table 5.5). ANOVA

demonstrated significant effects for treatment (dose nested in compound) [F8,88 = 9.612, p

< 0.001]. Mice treated with salvinorin A showed significant increases in null activity with

0.5 mg/kg (p < 0.001) compared to all other doses of the compound and the vehicle

controls. Effects of RB-64 were somewhat different. The two lower doses of RB-64

exerted little effects on null activity, whereas the two highest doses resulted in increased

activity compared to the vehicle control (p < 0.001).

111

Table 5.5 Null activity levels (mAmp displacement) in mice treated with vehicle, salvinorin A, or RB-64. Treatment Null Activity (mAmp displacement)

10% Tween Vehicle 11.16 ± 4.34

salvinorin A

0.25 mg/kg 19.44 ± 5.83

0.5 mg/kg 28.79 ± 3.99*

1.0 mg/kg 13.63 ± 3.65

2.0 mg/kg 13.76 ± 3.65

RB-64

0.005 mg/kg 8.22 ± 3.73

0.01 mg/kg 16.81 ± 4.4

0.05 mg/kg 64.89 ± 10.88*

0.1 mg/kg 51.6 ± 12.16*

*p < 0.05 compared to vehicle controls.

Baseline Startle Activity — Mice treated with salvinorin A or RB-64 showed dose-

dependent parabolic changes in baseline startle activity (Figure 5.6a). ANOVA revealed

significant main effects of salvinorin A and RB-64 treatment [F8,88 = 13.197, p < 0.001].

Bonferroni corrected pair-wise comparisons confirmed the dose-dependent effects of

salvinorin A on startle activity. The 0.25 and 0.5 mg/kg doses produced slight changes in

baseline startle responses, whereas 1 mg/kg salvinorin A resulted in enhanced startle

responses relative to the vehicle control (p < 0.004) and 0.25 and 2 mg/kg salvinorin A(p

< 0.034). Mice given RB-64 also showed dose-dependent parabolic changes (Figure

112

5.6a). Bonferroni comparisons revealed that the lowest dose of RB-64 had no effects on the startle response, whereas the 0.01 mg/kg dose produced augmented startle responses compared to vehicle (p < 0.001) and all other doses of RB-64 (p < 0.002).

PPI — Overall PPI was affected by salvinorin A (Figure 5.6b) and RB-64 (Figure 5.6C).

ANOVA yielded significant main effects of treatment for overall PPI [F8,88 = 6.606, p <

0.001]. Bonferroni corrected pair-wise comparisons found that overall inhibition was

similar for vehicle and 0.25, 0.5, or 1 mg/kg salvinorin-A, while 2 mg/kg significantly

suppressed inhibition (p < 0.044). By comparison, 0.01 and 0.05 mg/kg RB-64 resulted in

marked increases in overall inhibition compared to the vehicle control (p < 0.021). Mice

given 0.1 mg/kg RB-64 showed a significant reduction in overall inhibition relative to

vehicle-treated mice (p < 0.046).

Mice treated with vehicle showed prepulse-dependent PPI. Administration of salvinorin

A (Figure 5.6c) or RB-64 (Figure 5.6c) resulted in a loss of prepulse-dependent PPI and

the highest dose of each compound decreased overall inhibition. RMANOVA

demonstrated significant within-subject effects of prepulse intensity [F3,264 = 30.084, p <

0.001] and a significant prepulse intensity by treatment interaction [F24,264 = 4.602, p <

0.001]. Bonferroni corrected pair-wise comparisons confirmed prepulse-dependent PPI in

vehicle controls, with all prepulse intensities statistically different from each other (p <

0.011). Salvinorin A or RB-64 disrupted prepulse-dependent PPI at all prepulse

intensities. Mice given 0.25, 0.5 or 1 mg/kg salvinorin A had increased PPI to the 4 dB

prepulse relative to vehicle (p < 0.051) (Figure 5.6b); PPI responses at the 8 and 12 dB

113 prepulses were similar for animals administered vehicle, or 0.25, 0.5, or 1 mg/kg salvinorin A. However, mice given 2 mg/kg salvinorin A had marked reductions in PPI at the 8, 12, and 16 dB prepulses compared to vehicle (p > 0.011) or all other doses of the compound (p < 0.047); responses to 4 dB were unaffected from the vehicle control.

When animals were given 0.005, 0.01 or 0.05 mg/kg RB-64, PPI to the 4 dB prepulse were enhanced relative to vehicle (p < 0.047) (Figure 5.6c). By comparison, 0.005 RB-64 depressed responses to the 12 and 16 db prepulses (p < 0.010). The 0.01 and 0.05 doses increased PPI to the 8 dB prepulse relative to the vehicle controls (p < 0.031). Similarly,

0.05 mg/kg RB-64 augmented responses to the 12 and 16 db prepulses (p < 0.052).

Responses to the highest does of RB-64 were different than those for the 0.01 and 0.05 doses. Here, responses to the 12 and 16 dB prepulses were significantly depressed compared to the vehicle control (p < 0.003).

Collectively, these data show that PPI responses to salvinorin A and RB-64 show some differences and some similarities. Overall PPI was reduced by the highest dose of salvinorin A, while 0.01 and 0.05 mg/kg RB-64 increased and 0.1 mg/kg RB-64 decreased PPI. By comparison, both salvinorin A and RB-64 disrupted prepulse- dependent PPI.

114

Figure 5.6 PPI responses to salvinorin A and RB-64 by C57BL/6J mice. (6a) Startle responses to the 120 db stimulus for animals administered different doses of salvinorin Aor RB-64. (6b) Percent PPI to the 4, 8, 12, and 16 db prepulses for animals given various doses of salvinorin-A. (6c) Percent PPI to the 4, 8, 12, and 16 db prepulses for animals given various doses of RB-64. N = 10-16 mice/treatment. * p < 0.05, compared to vehicle controls; + p < 0.05, comparisons of prepulse-dependent PPI at the 4, 8, 12, and 16 dB prepulses within a single treatment; # p < 0.05, comparisons of startle responses (panel A) within compound to 0.5 mg/kg salvinorin A or 0.01 mg/kg RB-64.

5.3 Discussion

GPCRs are integral membrane proteins, which are well-known for their hydrophobic

nature. GPCRs and their derived peptides tend to aggregate in solution, bind to the container wall and segregate from the matrix. It is still a big challenge to structurally characterize GPCR proteins by MS and only limited cases are reported (242-245). A

115 previous attempt by Dr. Howells’s group to characterize KOR using trypsin in-gel digesion and MALDI-TOF MS only yielded 26% coverage (Poster presented in SFN meeting 2006). Similar low coverage was also seen in other GPCR studies, including

CB1 (35%) and CB2 (29%) (242), MOR (37%) and DOR (28%) (245, 246). The accessible peptides are usually located on N- and C-terminus, intracellular and extracellular loops, while the major TM domains are largely invisible. Our primary goal for the mass spectrometry study here was to identify the labeling site of RB-64. Matrix assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry was used to in our research due to its high sensitivity and tolerance to contaminants (e.g. salts). In our study, we also observed a low coverage 19-26% of total KOR sequence; however, a significant part of the accessible peptides were derived from TM domains

(Figure 5.4), including the RB-64 labeling site on TM7 (YFCIALGY). Our ongoing work involves using LC-MS to improve the coverage of total KOR sequence. The combination of high performance liquid chromatography (HPLC) with mass spectrometry could help separate the complex peptide sample and decrease the suppression of ion signal. In a recent MS study of receptor (CB2), nanoLC combined with LTQ-FT and the

4000 Q-Trap mass spectrometers achieved ~90% coverage of CB2 sequence (242).

For our sample preparation, we used a tandem affinity purification approach and then applied an in-gel chymotrypsin digestion of KOR. For complex protein mixtures, gel- based electrophoresis is often a favorable choice due to its high capacity for protein separation (247). In-gel digestion usually produces specific peptide fragments that are detectable by the mass analyzer. Every protein has a unique set of peptides digested by an

116 (e.g. trypsin or chymotrypsin), leading to unique peptide mass pattern. Trypsin is the most commonly used protease, cutting the protein at the carboxyl side of R and K residues. However, for our KOR protein, trypsin will yield a very large peptide (MW

7039.7 Da), which falls out of the MS detection range (600~4000 m/z). Chymotrypsin is a better choice for KOR because it cleaves at the C-terminus of (F), tyrosine (Y), (W), and (L) when these residues are not followed by proline (P). Moreover, these hydrophobic residues are richly distributed in TM domains.

In this approach, we do take the risk that chymotrypsin can cut our target peptide into very small pieces (CIALGY, MW 639.3); however a modified Cys has prevented the cleavage of nearby Y and F residues. So two-missed-cleavage peptide (YFCIALGY) within TM7 was the major product. It was clearly shown that this modified peptide (MW

1381.8) has a MW shift 431.5 as compared to the corresponding unmodified one (MW

949.5) (Table 5.4). Our ongoing research is to further confirm the labeling site by

MS/MS sequencing of YFCIAGY.

Besides the mass spectrometric analysis of the labeling site of RB-64, we also applied traditional pharmacological approach to quantify the labeling effects of RB-64 and other compounds (Figure 5.2). Diffusible ligands bind to receptors mainly through non- covalent interactions, such as electrostatic forces, hydrogen bonds, van der Waals’ forces and hydrophobic effects. In contrast, covalently-bound ligands can directly link to the receptor protein, thus having unique properties and pharmacological profiles. As is shown in our research (Table 5.1), RB-64 and RB-48 both have high affinities, potencies and selectivity to KOR. Furthermore, under certain conditions, both compounds can

117 irreversibly bind to the KOR by direct covalent labeling (Figure 5.2). In solution, these two compounds can undergo different chemical reaction pathways, namely addition and substitution reactions. However, in a biological environment, such as the binding pocket of KOR, multiple unknown factors (e.g. ligand orientation, side chain reactivity, etc.) may not allow all the possible chemical reactions to occur. Our mass spectrometric analysis of the affinity labeling clearly showed that both compounds underwent substitution reaction with only one MW shift (431.5) present. It means both RB-64 and

RB-48 underwent the substitution reaction, even though the isothiocyanate group in RB-

64 was a favorable group for addition reaction.

Mass spectrometry is a powerful tool to study posttranslational modification of proteins

(248). In our case, MS unambiguously identified the labeling site of covalently-bound ligands RB-64 and RB-48. Even though there are only a limited number of GPCRs have been studied by MS approach (242, 249-251), we are optimistic about the future of applying MS to structurally characterize GPCR proteins. The results obtained here validated our previous binding site model (2, 7, 168) and provided us insights into the mechanism for covalent modification of GPCRs. As a covalent binding ligand, RB-64 had shown great potential for being a molecular probe to explore opioid receptors, also for being a therapeutic agent due to its psychoactive nature.

118

Chapter 6: Future Directions and Conclusions

6.1 Future Directions

6.1.1 Quantifying G protein Expression

For our SCAM study (Chapter 4), we used G protein overexpression (2) to induce the conformational changes of KOR. Although the semi-quantification of G protein by antibody-blotting demonstrated an enhancement of the G protein-expression, the accurate

G protein quantity remains elusive. Here I propose to use an MS-based Stable

Labeling by Amino acids in Cell culture (SILAC) method to further quantify the G protein expression in HEK293 T, Gα16 and Gαi2 stably expressed cells as described in

chapter 4. For classic SILAC (Figure 6.1), cells are grown in culture medium containing

13C-labelled frequently occurring amino acids, such as Arg and Lys (252, 253). Generally

the isotope labeling requires a +4Da MW shift to avoid the undesirable overlap with the

13 naturally occurring isotopic patterns from individual peptides. The use of C6 -Arg can give rise to a +6 Da MW shift and allow a safer assignment of a labeled peptide as compared with unlabeled one. Recently differential labeled Arg permits the comparison of three different cells in one experiment (254, 255). This new technique uses natural

12 14 13 14 13 15 C6 - N4-Arg, isotope labeled C6 - N4-Arg (+6 Da), and C6 - N4-Arg (+10 Da), which will permit our study of G protein expression in three types of cells (HEK293 T,

Gα16 and Gαi2 stably expressed cells). During sample preparation, one type of cells is

supplied with one kind of differentially labeled Arg. Next, each type of cells is counted

and pooled together (1:1:1 ratio) before the lyses. Further purification (SDS-gel or LC)

can preserve the quantitative information in the original cell types since the differentially

119

labeled proteins go through the same purification, separation, and mass spectrometric

analysis process. The ratio of G protein is determined by examination of the MH+ ion

12 14 13 14 13 15 signal intensities which contain C6 - N4-Arg, C6 - N4-Arg (+6 Da), and C6 - N4-

Arg (+10 Da) in the full spectra.

MS

Figure 6.1 SILAC method for quantitative proteomics. Proteins are labeled metabolically by culturing cells in media that are isotopically enriched or isotopically depleted. Modified from Figure 3 of ref (252).

6.1.2 Chasing the Active and Global Conformation of KOR

My study of the KOR conformation mainly focused on TM7, the upper part of TM6 and

EL2. My colleague, Dr. Timothy A. Vortherms, conducted a similar SCAM study on

TM2 (160). These findings revealed a different pattern change for different TMs. TM 6 and 7 both showed a cluster pattern and there was a possible rotation of TM7. TM6 is well accepted being a toggle switch during the activation process of GPCR. TM2 didn’t show clearly the cluster pattern. However, SCAM can only give us an active-like conformation, not the true active conformation of KOR because there is no agonist

120 bound. One way to get an active conformation is to use our covalently-bound agonist,

RB-64, to label KOR, then perform the SCAM. The difference in the SCAM pattern between RB-64 labeled and unlabeled KOR can reveal some subtle changes in KOR conformation. Another way to get a global view of the conformational changes of KOR during the activation process is by NMR or MS. Both methods have been able to obtain high-resolution structural information on proteins and receptor-ligand interactions (256).

However, even with high-field magnets, selective labeling methods, and new pulse sequences, many proteins are too large for analysis by NMR. GPCR proteins are also difficult to be enriched to high concentrations required for NMR study. Here I propose to continue this conformation study by applying the MS-based Hydrogen/ exchange (H/D exchange) method. Amide H/D exchange is useful to study of protein structure and dynamics as well as receptor-ligand interactions (256-258). The degree to which an amide hydrogen is exposed to reveals the details of its environment.

Conformational changes induced by ligand binding cause variations in amide exposure, and these variations can be determined by measuring the rate of amide hydrogen exchange with solvent deuterium. Importantly, there is almost no limitation for protein concentrations used for H/D exchange. To compare the conformational changes during the activation process, H/D exchange experiments will be performed on both the free

KOR and KOR-RB-64 complex. The difference in exchange kinetics can reveal the precise information in receptor-ligand interactions. Mapping the whole protein by H/D exchange kinetics can thus identify ligand bound (active) and unbound (inactive-like)

KOR conformations.

121

6.1.3 Crystallographic Study of KOR

The crystallographic study of GPCR proteins still face many challenges (259), such as

GPCR expression, purification and crystallization. The conformational plasticity of

GPCRs makes the situation even worse. Up till now, only two GPCR structurs have been solved from high resolution X-ray data (17, 18, 25, 182). Among these available structures, rhodopsin has a high resolution inactive-state and a low resolution active-state structures and the β2AR has a high resolution inactive-like structure. The determination of an active-state structure would fill a large gap in our understandings of GPCR structure, and would provide more insight into GPCR function. Interestingly these two

GPCRs (rhdopsin and β2AR) were all co-crystallized with their corresponding ligands.

As we have detailed in chapter 5, the covalently-bound ligand RB-64, an agonist of KOR, provides an unprecedented opportunity for stabilizing KOR in active states as well as initiate crystal lattice formation. Our growing knowledge in KOR expression, purification and separation also added more hope for this challenging task.

6.2 General Conclusions

In this dissertation, I have detailed our study of the interactions between salvinorin A and

KOR through combined mutagenesis/computer modeling method. I also examined the

role that G protein overexpression played on KOR conformational changes. Our findings

revealed the molecular mechanism by which a small-molecule, salvinorin A, binds to and

activates KOR. Salvinorin A not only uses the majority of its flexible functional groups

(C-2, C-4 and C-12 functional groups) to have the optimal interactions with KOR, thus

122 stabilizing itself in the binding site with about -11 kCal/mol, but also takes advantage of the conformational changes induced by G protein-coupling (mainly Gαi2) which leads to

active state stabilization and activation of a series of downstream signaling events. GPCR

signaling seemingly is not only regulated by the receptor itself and its agonist, but also by the cellular environment in the case of our data with G proteins. All these observations

point to another level of complexity for GPCR signaling. Meanwhile, our rational design

of covalently-bound ligands based on salvinorin A’s structure further validated our

understanding of KOR and salvinorin A interactions. As a molecular probe, the covalent

ligand (RB-64) provides us with unprecedented opportunities to elucidate the structure

and function of KOR. It is our hope that a better understanding of opioid receptors, and

all other GPCRs, will facilitate advances in drug developments and greatly improve

human life.

123

Bibliography

1. Roth, B. L., Baner, K., Westkaemper, R., Siebert, D., Rice, K. C., Steinberg, S., Ernsberger, P., and Rothman, R. B. (2002) Salvinorin A: a potent naturally occurring nonnitrogenous kappa opioid selective agonist, Proc Natl Acad Sci U S A 99, 11934-11939. 2. Yan, F., Mosier, P. D., Westkaemper, R. B., Stewart, J., Zjawiony, J. K., Vortherms, T. A., Sheffler, D. J., and Roth, B. L. (2005) Identification of the Molecular Mechanisms by Which the Diterpenoid Salvinorin A Binds to κ-Opioid Receptors, Biochemistry 44, 8643-8651. 3. Bikbulatov, R. V., Stewart, J., Jin, W., Yan, F., Roth, B. L., Ferreira, D., and Zjawiony, J. K. (2008) Short synthesis of a novel class of salvinorin A analogs with hemiacetalic structure, Tetrahedron Letters 49, 937-940. 4. Bikbulatov, R. V., Yan, F., Roth, B. L., and Zjawiony, J. K. (2007) Convenient synthesis and in vitro pharmacological activity of 2-thioanalogs of salvinorins A and B, Bioorg Med Chem Lett 17, 2229-2232. 5. Munro, T. A., Rizzacasa, M. A., Roth, B. L., Toth, B. A., and Yan, F. (2005) Studies toward the pharmacophore of salvinorin A, a potent kappa opioid receptor agonist, J Med Chem 48, 345-348. 6. Stewart, D. J., Fahmy, H., Roth, B. L., Yan, F., and Zjawiony, J. K. (2006) Bioisosteric modification of salvinorin A, a potent and selective kappa-opioid receptor agonist, Arzneimittelforschung 56, 269-275. 7. Yan, F., Mosier, P. D., Westkaemper, R. B., and Roth, B. L. (2008) Galpha- Subunits Differentially Alter the Conformation and Agonist Affinity of kappa- Opioid Receptors, Biochemistry 47, 1567-1578. 8. Foord, S. M., Bonner, T. I., Neubig, R. R., Rosser, E. M., Pin, J. P., Davenport, A. P., Spedding, M., and Harmar, A. J. (2005) International Union of Pharmacology. XLVI. G protein-coupled receptor list, Pharmacol Rev 57, 279-288. 9. Zhang, Y., Devries, M. E., and Skolnick, J. (2006) Structure modeling of all identified G protein-coupled receptors in the , PLoS Comput Biol 2, e13. 10. Park, P. S., Lodowski, D. T., and Palczewski, K. (2007) Activation of G Protein- Coupled Receptors: Beyond Two-State Models and Tertiary Conformational Changes, Annu Rev Pharmacol Toxicol. 11. Venter, J. C., Adams, M. D., Myers, E. W., Li, P. W., Mural, R. J., Sutton, G. G., Smith, H. O., Yandell, M., Evans, C. A., Holt, R. A., Gocayne, J. D., Amanatides, P., Ballew, R. M., Huson, D. H., Wortman, J. R., Zhang, Q., Kodira, C. D., Zheng, X. H., Chen, L., Skupski, M., Subramanian, G., Thomas, P. D., Zhang, J., Gabor Miklos, G. L., Nelson, C., Broder, S., Clark, A. G., Nadeau, J., McKusick, V. A., Zinder, N., Levine, A. J., Roberts, R. J., Simon, M., Slayman, C., Hunkapiller, M., Bolanos, R., Delcher, A., Dew, I., Fasulo, D., Flanigan, M., Florea, L., Halpern, A., Hannenhalli, S., Kravitz, S., Levy, S., Mobarry, C., Reinert, K., Remington, K., Abu-Threideh, J., Beasley, E., Biddick, K., Bonazzi, V., Brandon, R., Cargill, M., Chandramouliswaran, I., Charlab, R., Chaturvedi, K., Deng, Z., Di Francesco, V., Dunn, P., Eilbeck, K., Evangelista, C., Gabrielian,

124

A. E., Gan, W., Ge, W., Gong, F., Gu, Z., Guan, P., Heiman, T. J., Higgins, M. E., Ji, R. R., Ke, Z., Ketchum, K. A., Lai, Z., Lei, Y., Li, Z., Li, J., Liang, Y., Lin, X., Lu, F., Merkulov, G. V., Milshina, N., Moore, H. M., Naik, A. K., Narayan, V. A., Neelam, B., Nusskern, D., Rusch, D. B., Salzberg, S., Shao, W., Shue, B., Sun, J., Wang, Z., Wang, A., Wang, X., Wang, J., Wei, M., Wides, R., Xiao, C., Yan, C., Yao, A., Ye, J., Zhan, M., Zhang, W., Zhang, H., Zhao, Q., Zheng, L., Zhong, F., Zhong, W., Zhu, S., Zhao, S., Gilbert, D., Baumhueter, S., Spier, G., Carter, C., Cravchik, A., Woodage, T., Ali, F., An, H., Awe, A., Baldwin, D., Baden, H., Barnstead, M., Barrow, I., Beeson, K., Busam, D., Carver, A., Center, A., Cheng, M. L., Curry, L., Danaher, S., Davenport, L., Desilets, R., Dietz, S., Dodson, K., Doup, L., Ferriera, S., Garg, N., Gluecksmann, A., Hart, B., Haynes, J., Haynes, C., Heiner, C., Hladun, S., Hostin, D., Houck, J., Howland, T., Ibegwam, C., Johnson, J., Kalush, F., Kline, L., Koduru, S., Love, A., Mann, F., May, D., McCawley, S., McIntosh, T., McMullen, I., Moy, M., Moy, L., Murphy, B., Nelson, K., Pfannkoch, C., Pratts, E., Puri, V., Qureshi, H., Reardon, M., Rodriguez, R., Rogers, Y. H., Romblad, D., Ruhfel, B., Scott, R., Sitter, C., Smallwood, M., Stewart, E., Strong, R., Suh, E., Thomas, R., Tint, N. N., Tse, S., Vech, C., Wang, G., Wetter, J., Williams, S., Williams, M., Windsor, S., Winn- Deen, E., Wolfe, K., Zaveri, J., Zaveri, K., Abril, J. F., Guigo, R., Campbell, M. J., Sjolander, K. V., Karlak, B., Kejariwal, A., Mi, H., Lazareva, B., Hatton, T., Narechania, A., Diemer, K., Muruganujan, A., Guo, N., Sato, S., Bafna, V., Istrail, S., Lippert, R., Schwartz, R., Walenz, B., Yooseph, S., Allen, D., Basu, A., Baxendale, J., Blick, L., Caminha, M., Carnes-Stine, J., Caulk, P., Chiang, Y. H., Coyne, M., Dahlke, C., Mays, A., Dombroski, M., Donnelly, M., Ely, D., Esparham, S., Fosler, C., Gire, H., Glanowski, S., Glasser, K., Glodek, A., Gorokhov, M., Graham, K., Gropman, B., Harris, M., Heil, J., Henderson, S., Hoover, J., Jennings, D., Jordan, C., Jordan, J., Kasha, J., Kagan, L., Kraft, C., Levitsky, A., Lewis, M., Liu, X., Lopez, J., Ma, D., Majoros, W., McDaniel, J., Murphy, S., Newman, M., Nguyen, T., Nguyen, N., Nodell, M., Pan, S., Peck, J., Peterson, M., Rowe, W., Sanders, R., Scott, J., Simpson, M., Smith, T., Sprague, A., Stockwell, T., Turner, R., Venter, E., Wang, M., Wen, M., Wu, D., Wu, M., Xia, A., Zandieh, A., and Zhu, X. (2001) The sequence of the human genome, Science 291, 1304-1351. 12. Kroeze, W. K., Sheffler, D. J., and Roth, B. L. (2003) G-protein-coupled receptors at a glance, J Cell Sci 116, 4867-4869. 13. (2004) Finishing the euchromatic sequence of the human genome, Nature 431, 931-945. 14. Overington, J. P., Al-Lazikani, B., and Hopkins, A. L. (2006) How many drug targets are there?, Nat Rev Drug Discov 5, 993-996. 15. Jacoby, E., Bouhelal, R., Gerspacher, M., and Seuwen, K. (2006) The 7 TM G- protein-coupled receptor target family, ChemMedChem 1, 761-782. 16. Henderson, R., and Unwin, P. N. (1975) Three-dimensional model of purple membrane obtained by electron microscopy, Nature 257, 28-32. 17. Palczewski, K., Kumasaka, T., Hori, T., Behnke, C. A., Motoshima, H., Fox, B. A., Le Trong, I., Teller, D. C., Okada, T., Stenkamp, R. E., Yamamoto, M., and

125

Miyano, M. (2000) Crystal structure of rhodopsin: A G protein-coupled receptor, Science 289, 739-745. 18. Salom, D., Lodowski, D. T., Stenkamp, R. E., Le Trong, I., Golczak, M., Jastrzebska, B., Harris, T., Ballesteros, J. A., and Palczewski, K. (2006) Crystal structure of a photoactivated deprotonated intermediate of rhodopsin, Proc Natl Acad Sci U S A 103, 16123-16128. 19. Filipek, S., Stenkamp, R. E., Teller, D. C., and Palczewski, K. (2003) G protein- coupled receptor rhodopsin: a prospectus, Annu Rev Physiol 65, 851-879. 20. Filipek, S., Teller, D. C., Palczewski, K., and Stenkamp, R. (2003) The crystallographic model of rhodopsin and its use in studies of other G protein- coupled receptors, Annu Rev Biophys Biomol Struct 32, 375-397. 21. Fotiadis, D., Jastrzebska, B., Philippsen, A., Muller, D. J., Palczewski, K., and Engel, A. (2006) Structure of the rhodopsin dimer: a working model for G- protein-coupled receptors, Curr Opin Struct Biol 16, 252-259. 22. Palczewski, K. (2006) G protein-coupled receptor rhodopsin, Annu Rev Biochem 75, 743-767. 23. Salom, D., Lodowski, D. T., Stenkamp, R. E., Le Trong, I., Golczak, M., Jastrzebska, B., Harris, T., Ballesteros, J. A., and Palczewski, K. (2006) Crystal Structure of a Photoactivated Deprotonated Intermediate of Rhodopsin, Proc. Nat. Acad. Sci. USA 103, 16123-16128. 24. Rosenbaum, D. M., Cherezov, V., Hanson, M. A., Rasmussen, S. G., Thian, F. S., Kobilka, T. S., Choi, H. J., Yao, X. J., Weis, W. I., Stevens, R. C., and Kobilka, B. K. (2007) GPCR engineering yields high-resolution structural insights into beta2-adrenergic receptor function, Science 318, 1266-1273. 25. Cherezov, V., Rosenbaum, D. M., Hanson, M. A., Rasmussen, S. G., Thian, F. S., Kobilka, T. S., Choi, H. J., Kuhn, P., Weis, W. I., Kobilka, B. K., and Stevens, R. C. (2007) High-resolution crystal structure of an engineered human beta2- adrenergic G protein-coupled receptor, Science 318, 1258-1265. 26. Day, P. W., Rasmussen, S. G., Parnot, C., Fung, J. J., Masood, A., Kobilka, T. S., Yao, X. J., Choi, H. J., Weis, W. I., Rohrer, D. K., and Kobilka, B. K. (2007) A monoclonal antibody for G protein-coupled receptor crystallography, Nat Methods 4, 927-929. 27. Rasmussen, S. G., Choi, H. J., Rosenbaum, D. M., Kobilka, T. S., Thian, F. S., Edwards, P. C., Burghammer, M., Ratnala, V. R., Sanishvili, R., Fischetti, R. F., Schertler, G. F., Weis, W. I., and Kobilka, B. K. (2007) Crystal structure of the human beta(2) adrenergic G-protein-coupled receptor, Nature. 28. Rovati, G. E., Capra, V., and Neubig, R. R. (2007) The highly conserved DRY motif of class A G protein-coupled receptors: beyond the ground state, Mol Pharmacol 71, 959-964. 29. Christopoulos, A., and Kenakin, T. (2002) G protein-coupled receptor allosterism and complexing, Pharmacol Rev 54, 323-374. 30. Schwartz, T. W., Frimurer, T. M., Holst, B., Rosenkilde, M. M., and Elling, C. E. (2006) Molecular mechanism of 7TM receptor activation--a global toggle switch model, Annu Rev Pharmacol Toxicol 46, 481-519. 31. Downes, G. B., and Gautam, N. (1999) The G protein subunit gene families, Genomics 62, 544-552.

126

32. Gilman, A. G. (1987) G proteins: transducers of receptor-generated signals, Annu Rev Biochem 56, 615-649. 33. Simonds, W. F. (1999) G protein regulation of adenylate cyclase Trends Pharmacol Sci 20, 66-73. 34. Sternweis, P. C. (1994) The active role of beta gamma in signal transduction, Curr Opin Cell Biol 6, 198-203. 35. Dupre, D. J., and Hebert, T. E. (2006) and trafficking of seven transmembrane receptor signalling complexes, Cell Signal 18, 1549-1559. 36. Kholodenko, B. N. (2006) Cell-signalling dynamics in time and space, Nat Rev Mol Cell Biol 7, 165-176. 37. Ferguson, S. S. (2001) Evolving concepts in G protein-coupled receptor endocytosis: the role in receptor desensitization and signaling, Pharmacol Rev 53, 1-24. 38. Ferguson, S. S., Zhang, J., Barak, L. S., and Caron, M. G. (1998) Molecular mechanisms of G protein-coupled receptor desensitization and resensitization, Life Sci 62, 1561-1565. 39. DeWire, S. M., Ahn, S., Lefkowitz, R. J., and Shenoy, S. K. (2007) Beta-arrestins and cell signaling, Annu Rev Physiol 69, 483-510. 40. Lefkowitz, R. J., Rajagopal, K., and Whalen, E. J. (2006) New roles for beta- arrestins in cell signaling: not just for seven-transmembrane receptors, Mol Cell 24, 643-652. 41. Violin, J. D., and Lefkowitz, R. J. (2007) Beta-arrestin-biased ligands at seven- transmembrane receptors, Trends Pharmacol Sci 28, 416-422. 42. Lefkowitz, R. J., and Shenoy, S. K. (2005) Transduction of receptor signals by beta-arrestins, Science 308, 512-517. 43. Kirkham, M., and Parton, R. G. (2005) Clathrin-independent endocytosis: new insights into caveolae and non-caveolar lipid raft carriers, Biochim Biophys Acta 1746, 349-363. 44. Hommelgaard, A. M., Roepstorff, K., Vilhardt, F., Torgersen, M. L., Sandvig, K., and van Deurs, B. (2005) Caveolae: stable membrane domains with a potential for internalization, Traffic 6, 720-724. 45. Parton, R. G., and Simons, K. (2007) The multiple faces of caveolae, Nat Rev Mol Cell Biol 8, 185-194. 46. Kenakin, T. (2004) Principles: receptor theory in pharmacology, Trends Pharmacol Sci 25, 186-192. 47. Maehle, A. H., Prull, C. R., and Halliwell, R. F. (2002) The emergence of the drug receptor theory, Nat Rev Drug Discov 1, 637-641. 48. Limbird, L. E. (2004) The receptor concept: a continuing evolution, Mol Interv 4, 326-336. 49. Whorton, M. R., Bokoch, M. P., Rasmussen, S. G., Huang, B., Zare, R. N., Kobilka, B., and Sunahara, R. K. (2007) A monomeric G protein-coupled receptor isolated in a high-density lipoprotein particle efficiently activates its G protein, Proc Natl Acad Sci U S A 104, 7682-7687. 50. De , A., Stadel, J. M., and Lefkowitz, R. J. (1980) A ternary complex model explains the agonist-specific binding properties of the adenylate cyclase-coupled beta-adrenergic receptor, J Biol Chem 255, 7108-7117.

127

51. Duchene, J., Chauhan, S. D., Lopez, F., Pecher, C., Esteve, J. P., Girolami, J. P., Bascands, J. L., and Schanstra, J. P. (2005) Direct protein-protein interaction between PLCgamma1 and the B2 receptor--importance of growth conditions, Biochem Biophys Res Commun 326, 894-900. 52. Gomes, I., Jordan, B. A., Gupta, A., Trapaidze, N., Nagy, V., and Devi, L. A. (2000) Heterodimerization of mu and delta opioid receptors: A role in synergy, J Neurosci 20, RC110. 53. White, J. F., Grodnitzky, J., Louis, J. M., Trinh, L. B., Shiloach, J., Gutierrez, J., Northup, J. K., and Grisshammer, R. (2007) Dimerization of the class A G protein-coupled receptor NTS1 alters G protein interaction, Proc Natl Acad Sci U S A 104, 12199-12204. 54. Kobilka, B. K., and Deupi, X. (2007) Conformational complexity of G-protein- coupled receptors, Trends Pharmacol Sci 28, 397-406. 55. McCusker, E. C., Bane, S. E., O'Malley, M. A., and Robinson, A. S. (2007) Heterologous GPCR expression: a bottleneck to obtaining crystal structures, Biotechnol Prog 23, 540-547. 56. Maudsley, S., Martin, B., and Luttrell, L. M. (2007) G protein-coupled receptor signaling complexity in neuronal tissue: implications for novel therapeutics, Curr Alzheimer Res 4, 3-19. 57. Dorsam, R. T., and Gutkind, J. S. (2007) G-protein-coupled receptors and cancer, Nat Rev Cancer 7, 79-94. 58. SHEFFLER, D. J. (2006) THE REGULATION OF G PROTEIN-COUPLED RECEPTOR (GPCR) SIGNAL TRANSDUCTION BY p90 RIBOSOMAL S6 KINASE 2 (RSK2), Ph.D Thesis (Case Western Reserve University). 59. Eguchi, M. (2004) Recent advances in selective opioid receptor agonists and antagonists, Med Res Rev 24, 182-212. 60. Janecka, A., Fichna, J., and Janecki, T. (2004) Opioid receptors and their ligands, Curr Top Med Chem 4, 1-17. 61. Minami, M., and Satoh, M. (1995) Molecular biology of the opioid receptors: structures, functions and distributions, Neurosci Res 23, 121-145. 62. Satoh, M., and Minami, M. (1995) Molecular pharmacology of the opioid receptors, Pharmacol Ther 68, 343-364. 63. Gilbert, P. E., and Martin, W. R. (1976) The effects of morphine and nalorphine- like drugs in the nondependent, morphine-dependent and -dependent chronic spinal dog, J Pharmacol Exp Ther 198, 66-82. 64. Lord, J. A., Waterfield, A. A., Hughes, J., and Kosterlitz, H. W. (1977) Endogenous opioid peptides: multiple agonists and receptors, Nature 267, 495- 499. 65. Henderson, G., and McKnight, A. T. (1997) The orphan opioid receptor and its endogenous ligand--nociceptin/orphanin FQ, Trends Pharmacol Sci 18, 293-300. 66. Wang, J. B., Johnson, P. S., Persico, A. M., Hawkins, A. L., Griffin, C. A., and Uhl, G. R. (1994) Human mu opiate receptor. cDNA and genomic clones, pharmacologic characterization and chromosomal assignment, FEBS Lett 338, 217-222.

128

67. Knapp, R. J., Malatynska, E., Fang, L., Li, X., Babin, E., Nguyen, M., Santoro, G., Varga, E. V., Hruby, V. J., Roeske, W. R., and et al. (1994) Identification of a human delta opioid receptor: cloning and expression, Life Sci 54, PL463-469. 68. Simonin, F., Gaveriaux-Ruff, C., Befort, K., Matthes, H., Lannes, B., Micheletti, G., Mattei, M. G., Charron, G., Bloch, B., and Kieffer, B. (1995) kappa-Opioid receptor in humans: cDNA and genomic cloning, chromosomal assignment, functional expression, pharmacology, and expression pattern in the central nervous system, Proc Natl Acad Sci U S A 92, 7006-7010. 69. Simonin, F., Befort, K., Gaveriaux-Ruff, C., Matthes, H., Nappey, V., Lannes, B., Micheletti, G., and Kieffer, B. (1994) The human delta-opioid receptor: genomic organization, cDNA cloning, functional expression, and distribution in , Mol Pharmacol 46, 1015-1021. 70. Law, P. Y., Wong, Y. H., and Loh, H. H. (2000) Molecular mechanisms and regulation of opioid receptor signaling, Annu Rev Pharmacol Toxicol 40, 389- 430. 71. Corbett, A. D., Henderson, G., McKnight, A. T., and Paterson, S. J. (2006) 75 years of opioid research: the exciting but vain quest for the Holy Grail, Br J Pharmacol 147 Suppl 1, S153-162. 72. Waldhoer, M., Bartlett, S. E., and Whistler, J. L. (2004) Opioid receptors, Annu Rev Biochem 73, 953-990. 73. Dhawan, B. N., Cesselin, F., Raghubir, R., Reisine, T., Bradley, P. B., Portoghese, P. S., and Hamon, M. (1996) International Union of Pharmacology. XII. Classification of opioid receptors, Pharmacol Rev 48, 567-592. 74. Mansour, A., Fox, C. A., Akil, H., and Watson, S. J. (1995) Opioid-receptor mRNA expression in the rat CNS: anatomical and functional implications, Trends Neurosci 18, 22-29. 75. Bagnol, D., Mansour, A., Akil, H., and Watson, S. J. (1997) Cellular localization and distribution of the cloned mu and kappa opioid receptors in rat gastrointestinal tract, Neuroscience 81, 579-591. 76. Fickel, J., Bagnol, D., Watson, S. J., and Akil, H. (1997) Opioid receptor expression in the rat gastrointestinal tract: a quantitative study with comparison to the brain, Brain Res Mol Brain Res 46, 1-8. 77. Brown, D. R., Poonyachoti, S., Osinski, M. A., Kowalski, T. R., Pampusch, M. S., Elde, R. P., and Murtaugh, M. P. (1998) Delta-opioid receptor mRNA expression and immunohistochemical localization in porcine ileum, Dig Dis Sci 43, 1402-1410. 78. Peterson, P. K., Gekker, G., Lokensgard, J. R., Bidlack, J. M., Chang, A. C., Fang, X., and Portoghese, P. S. (2001) Kappa-opioid receptor agonist suppression of HIV-1 expression in CD4+ lymphocytes, Biochem Pharmacol 61, 1145-1151. 79. Greengard, P., Jen, J., Nairn, A. C., and Stevens, C. F. (1991) Enhancement of the glutamate response by cAMP-dependent protein kinase in hippocampal neurons, Science 253, 1135-1138. 80. Chavez-Noriega, L. E., and Stevens, C. F. (1994) Increased transmitter release at excitatory produced by direct activation of adenylate cyclase in rat hippocampal slices, J Neurosci 14, 310-317.

129

81. Williams, J. T., Christie, M. J., and Manzoni, O. (2001) Cellular and synaptic adaptations mediating opioid dependence, Physiol Rev 81, 299-343. 82. Clapham, D. E., and Neer, E. J. (1997) G protein beta gamma subunits, Annu Rev Pharmacol Toxicol 37, 167-203. 83. Wettschureck, N., and Offermanns, S. (2005) Mammalian G proteins and their cell type specific functions, Physiol Rev 85, 1159-1204. 84. Lee, J. W., Joshi, S., Chan, J. S., and Wong, Y. H. (1998) Differential coupling of mu-, delta-, and kappa-opioid receptors to G alpha16-mediated stimulation of phospholipase C, J Neurochem 70, 2203-2211. 85. Goldsmith, Z. G., and Dhanasekaran, D. N. (2007) G protein regulation of MAPK networks, Oncogene 26, 3122-3142. 86. Pierce, C. (2003) Drugs of abuse and the limbic system, Sigma RBI brochure. 87. Bailey, C. P., and Connor, M. (2005) Opioids: cellular mechanisms of tolerance and physical dependence, Curr Opin Pharmacol 5, 60-68. 88. Koch, T., Widera, A., Bartzsch, K., Schulz, S., Brandenburg, L. O., Wundrack, N., Beyer, A., Grecksch, G., and Hollt, V. (2005) Receptor endocytosis counteracts the development of opioid tolerance, Mol Pharmacol 67, 280-287. 89. Zhang, J., Ferguson, S. S., Barak, L. S., Bodduluri, S. R., Laporte, S. A., Law, P. Y., and Caron, M. G. (1998) Role for G protein-coupled receptor kinase in agonist-specific regulation of mu-opioid receptor responsiveness, Proc Natl Acad Sci U S A 95, 7157-7162. 90. Jordan, B. A., and Devi, L. A. (1999) G-protein-coupled receptor heterodimerization modulates receptor function, Nature 399, 697-700. 91. Cvejic, S., and Devi, L. A. (1997) Dimerization of the delta opioid receptor: implication for a role in receptor internalization, J Biol Chem 272, 26959-26964. 92. Terskiy, A., Wannemacher, K. M., Yadav, P. N., Tsai, M., Tian, B., and Howells, R. D. (2007) Search of the human for endomorphin-1 and endomorphin-2 precursor proteins, Life Sci 81, 1593-1601. 93. Hughes, J., Smith, T. W., Kosterlitz, H. W., Fothergill, L. A., Morgan, B. A., and Morris, H. R. (1975) Identification of two related pentapeptides from the brain with potent opiate agonist activity, Nature 258, 577-580. 94. Cox, B. M., Goldstein, A., and Hi, C. H. (1976) Opioid activity of a peptide, beta- -(61-91), derived from beta-lipotropin, Proc Natl Acad Sci U S A 73, 1821-1823. 95. Mollereau, C., Simons, M. J., Soularue, P., Liners, F., Vassart, G., Meunier, J. C., and Parmentier, M. (1996) Structure, tissue distribution, and chromosomal localization of the prepronociceptin gene, Proc Natl Acad Sci U S A 93, 8666- 8670. 96. Goldstein, A., Tachibana, S., Lowney, L. I., Hunkapiller, M., and Hood, L. (1979) Dynorphin-(1-13), an extraordinarily potent , Proc Natl Acad Sci U S A 76, 6666-6670. 97. Reinscheid, R. K., Nothacker, H. P., Bourson, A., Ardati, A., Henningsen, R. A., Bunzow, J. R., Grandy, D. K., Langen, H., Monsma, F. J., Jr., and Civelli, O. (1995) Orphanin FQ: a that activates an opioidlike G protein- coupled receptor, Science 270, 792-794. 98. Pawl, R. (2007) Opioids, Surg Neurol 68, 646-647.

130

99. Somogyi, A. A., Barratt, D. T., and Coller, J. K. (2007) Pharmacogenetics of opioids, Clin Pharmacol Ther 81, 429-444. 100. Sterken, J., Troubleyn, J., Gasthuys, F., Maes, V., Diltoer, M., and Verborgh, C. (2004) Intentional overdose of Large Animal Immobilon, Eur J Emerg Med 11, 298-301. 101. Houde, R. W. (1979) Analgesic effectiveness of the agonist-antagonists, Br J Clin Pharmacol 7 Suppl 3, 297S-308S. 102. Bowdle, T. A. (1998) Adverse effects of opioid agonists and agonist-antagonists in anaesthesia, Drug Saf 19, 173-189. 103. Ballantyne, J. C., and LaForge, K. S. (2007) Opioid dependence and addiction during opioid treatment of , Pain 129, 235-255. 104. Gold, M. S., Dackis, C. A., Pottash, A. L., Sternbach, H. H., Annitto, W. J., Martin, D., and Dackis, M. P. (1982) Naltrexone, opiate addiction, and , Med Res Rev 2, 211-246. 105. Calderon, S. N., and Coop, A. (2004) SNC 80 and related delta opioid agonists, Curr Pharm Des 10, 733-742. 106. Broom, D. C., Jutkiewicz, E. M., Rice, K. C., Traynor, J. R., and Woods, J. H. (2002) Behavioral effects of delta-opioid receptor agonists: potential ?, Jpn J Pharmacol 90, 1-6. 107. Jutkiewicz, E. M. (2006) The antidepressant -like effects of delta-opioid receptor agonists, Mol Interv 6, 162-169. 108. Yan, F., and Roth, B. L. (2004) Salvinorin A: a novel and highly selective kappa- opioid receptor agonist, Life Sci 75, 2615-2619. 109. Dortch-Carnes, J., and Potter, D. E. (2005) Bremazocine: a kappa-opioid agonist with potent analgesic and other pharmacologic properties, CNS Drug Rev 11, 195- 212. 110. Pande, A. C., Pyke, R. E., Greiner, M., Cooper, S. A., Benjamin, R., and Pierce, M. W. (1996) Analgesic efficacy of the kappa-receptor agonist, enadoline, in dental surgery pain, Clin Neuropharmacol 19, 92-97. 111. Peng, X., and Neumeyer, J. L. (2007) Kappa receptor bivalent ligands, Curr Top Med Chem 7, 363-373. 112. Machelska, H., Pfluger, M., Weber, W., Piranvisseh-Volk, M., Daubert, J. D., Dehaven, R., and Stein, C. (1999) Peripheral effects of the kappa-opioid agonist EMD 61753 on pain and inflammation in rats and humans, J Pharmacol Exp Ther 290, 354-361. 113. Eisenach, J. C., Carpenter, R., and Curry, R. (2003) Analgesia from a peripherally active kappa-opioid receptor agonist in patients with chronic pancreatitis, Pain 101, 89-95. 114. Millan, M. J. (1990) Kappa-opioid receptors and analgesia, Trends Pharmacol Sci 11, 70-76. 115. Ballantyne, J. C. (2007) Opioid misuse in oncology pain patients, Curr Pain Rep 11, 276-282. 116. Feldman, P. L., James, M. K., Brackeen, M. F., Bilotta, J. M., Schuster, S. V., Lahey, A. P., Lutz, M. W., Johnson, M. R., and Leighton, H. J. (1991) Design, synthesis, and pharmacological evaluation of ultrashort- to long-acting opioid analgetics, J Med Chem 34, 2202-2208.

131

117. Fitzgibbon, D. R. (2007) Clinical use of opioids for cancer pain, Curr Pain Headache Rep 11, 251-258. 118. Jones, S., and Bonci, A. (2005) Synaptic plasticity and drug addiction, Curr Opin Pharmacol 5, 20-25. 119. Manlandro, J. J., Jr. (2007) Using buprenorphine for outpatient opioid detoxification, J Am Osteopath Assoc 107, ES11-16. 120. Manlandro, J. J., Jr. (2005) Buprenorphine for office-based treatment of patients with opioid addiction, J Am Osteopath Assoc 105, S8-13. 121. Johnson, J. B. (1939) The elements of Mazatec witchcraft, Göteborgs Etnografiska Museum Etnologiska Studier 9, 119-149. 122. Wasson, R. G. (1962) A new Mexican psychotropic drug from the mint family, Botanical Museum Leaflets, Harvard University 20, 77-84. 123. Epling, C. J., M. C. (1962) A new species of salvia from Mexico, Botanical Museum Leaflets, Harvard University 20, 75-76. 124. Valdes, L. J., 3rd, Diaz, J. L., and Paul, A. G. (1983) Ethnopharmacology of ska Maria Pastora (Salvia divinorum, Epling and Jativa-M.), J Ethnopharmacol 7, 287-312. 125. Siebert, D. J. (1994) Salvia divinorum and salvinorin A: new pharmacologic findings, J Ethnopharmacol 43, 53-56. 126. Kacela, X. (2006) Being one with the spirit: dimensions of a mystical experience, J Pastoral Care Counsel 60, 83-94. 127. Griffiths, R. R., Richards, W. A., McCann, U., and Jesse, R. (2006) Psilocybin can occasion mystical-type experiences having substantial and sustained personal meaning and spiritual significance, Psychopharmacology (Berl) 187, 268-283; discussion 284-292. 128. Baker, J. R. (2005) Psychedelic sacraments, J Psychoactive Drugs 37, 179-187. 129. Alfredo Ortega, J. F. B. a. P. S. M. (1982) Salvinorin, a new trans-neoclerodane diterpene from Salvia divinorum(Labiatae), J. Chem. Soc., Perkin Trans. 1, 2505 - 2508. 130. Zhang, Y., Butelman, E. R., Schlussman, S. D., Ho, A., and Kreek, M. J. (2005) Effects of the plant-derived hallucinogen salvinorin A on basal dopamine levels in the caudate putamen and in a conditioned place aversion assay in mice: agonist actions at kappa opioid receptors, Psychopharmacology (Berl) 179, 551-558. 131. Prisinzano, T. E. (2005) Psychopharmacology of the hallucinogenic sage Salvia divinorum, Life Sci 78, 527-531. 132. McCurdy, C. R., Sufka, K. J., Smith, G. H., Warnick, J. E., and Nieto, M. J. (2006) Antinociceptive profile of salvinorin A, a structurally unique kappa opioid receptor agonist, Pharmacol Biochem Behav 83, 109-113. 133. Wang, Y., Tang, K., Inan, S., Siebert, D., Holzgrabe, U., Lee, D. Y., Huang, P., Li, J. G., Cowan, A., and Liu-Chen, L. Y. (2005) Comparison of pharmacological activities of three distinct kappa ligands (Salvinorin A, TRK-820 and 3FLB) on kappa opioid receptors in vitro and their antipruritic and antinociceptive activities in vivo, J Pharmacol Exp Ther 312, 220-230. 134. John, T. F., French, L. G., and Erlichman, J. S. (2006) The antinociceptive effect of Salvinorin A in mice, European Journal of Pharmacology 545, 129-133.

132

135. Capasso, R., Borrelli, F., Capasso, F., Siebert, D. J., Stewart, D. J., Zjawiony, J. K., and Izzo, A. A. (2006) The hallucinogenic herb Salvia divinorum and its active ingredient salvinorin A inhibit enteric cholinergic transmission in the guinea-pig ileum, Neurogastroenterol Motil 18, 69-75. 136. Rothman, R. B., Gorelick, D. A., Heishman, S. J., Eichmiller, P. R., Hill, B. H., Norbeck, J., and Liberto, J. G. (2000) An open-label study of a functional opioid kappa antagonist in the treatment of opioid dependence, J Subst Abuse Treat 18, 277-281. 137. Carlezon, W. A., Jr., Beguin, C., DiNieri, J. A., Baumann, M. H., Richards, M. R., Todtenkopf, M. S., Rothman, R. B., Ma, Z., Lee, D. Y., and Cohen, B. M. (2006) Depressive-like effects of the kappa-opioid receptor agonist salvinorin A on behavior and neurochemistry in rats, J Pharmacol Exp Ther 316, 440-447. 138. Todtenkopf, M. S., Marcus, J. F., Portoghese, P. S., and Carlezon, W. A., Jr. (2004) Effects of kappa-opioid receptor ligands on intracranial self-stimulation in rats, Psychopharmacology (Berl) 172, 463-470. 139. Mague, S. D., Pliakas, A. M., Todtenkopf, M. S., Tomasiewicz, H. C., Zhang, Y., Stevens, W. C., Jr., Jones, R. M., Portoghese, P. S., and Carlezon, W. A., Jr. (2003) Antidepressant-like effects of kappa-opioid receptor antagonists in the forced swim test in rats, J Pharmacol Exp Ther 305, 323-330. 140. Sheffler, D. J., and Roth, B. L. (2003) Salvinorin A: the "magic mint" hallucinogen finds a molecular target in the kappa opioid receptor, Trends Pharmacol Sci 24, 107-109. 141. Grundmann, O., Phipps, S. M., Zadezensky, I., and Butterweck, V. (2007) Salvia divinorum and salvinorin A: an update on pharmacology and analytical methodology, Planta Med 73, 1039-1046. 142. Ma, J., Ye, N., Lange, N., and Cohen, B. M. (2003) Dynorphinergic GABA neurons are a target of both typical and drugs in the nucleus accumbens shell, central amygdaloid nucleus and thalamic central medial nucleus, Neuroscience 121, 991-998. 143. Gulya, K. (1990) The opioid system in neurologic and psychiatric disorders and in their experimental models, Pharmacol Ther 46, 395-428. 144. Vortherms, T. A., and Roth, B. L. (2006) Salvinorin A: from to human therapeutics, Mol Interv 6, 257-265. 145. Mowry, M., Mosher, M., and Briner, W. (2003) Acute physiologic and chronic histologic changes in rats and mice exposed to the unique hallucinogen salvinorin A, J Psychoactive Drugs 35, 379-382. 146. Butelman, E. R., Harris, T. J., and Kreek, M. J. (2004) The plant-derived hallucinogen, salvinorin A, produces kappa-opioid agonist-like discriminative effects in rhesus monkeys, Psychopharmacology (Berl) 172, 220-224. 147. Schmidt, M. D., Schmidt, M. S., Butelman, E. R., Harding, W. W., Tidgewell, K., Murry, D. J., Kreek, M. J., and Prisinzano, T. E. (2005) of the plant-derived kappa-opioid hallucinogen salvinorin A in nonhuman primates, 58, 208-210. 148. Fantegrossi, W. E., Kugle, K. M., Valdes, L. J., 3rd, Koreeda, M., and Woods, J. H. (2005) Kappa-opioid receptor-mediated effects of the plant-derived

133

hallucinogen, salvinorin A, on inverted screen performance in the mouse, Behav Pharmacol 16, 627-633. 149. John, T. F., French, L. G., and Erlichman, J. S. (2006) The antinociceptive effect of salvinorin A in mice, Eur J Pharmacol 545, 129-133. 150. Willmore-Fordham, C. B., Krall, D. M., McCurdy, C. R., and Kinder, D. H. (2007) The hallucinogen derived from Salvia divinorum, salvinorin A, has kappa- opioid agonist discriminative stimulus effects in rats, 53, 481-486. 151. Braida, D., Limonta, V., Pegorini, S., Zani, A., Guerini-Rocco, C., Gori, E., and Sala, M. (2007) Hallucinatory and rewarding effect of salvinorin A in zebrafish: kappa-opioid and CB1- involvement, Psychopharmacology (Berl) 190, 441-448. 152. Ansonoff, M. A., Zhang, J., Czyzyk, T., Rothman, R. B., Stewart, J., Xu, H., Zjwiony, J., Siebert, D. J., Yang, F., Roth, B. L., and Pintar, J. E. (2006) Antinociceptive and hypothermic effects of Salvinorin A are abolished in a novel strain of kappa-opioid receptor-1 knockout mice, J Pharmacol Exp Ther 318, 641-648. 153. Beeler, A. B., Acquilano, D. E., Su, Q., Yan, F., Roth, B. L., Panek, J. S., and Porco, J. A., Jr. (2005) Synthesis of a library of complex macrodiolides employing cyclodimerization of hydroxy esters, J Comb Chem 7, 673-681. 154. Setola, V. (2005) Discovering the molecular and cellular mechanisms underlying -induced cardiopulmonary side effects, pp xvii, 170 p., Case Western Reserve University, Cleveland, Ohio. 155. Kozak, M. (1994) Features in the 5' non-coding sequences of rabbit alpha and beta-globin mRNAs that affect translational efficiency, J Mol Biol 235, 95-110. 156. Xu, W., Campillo, M., Pardo, L., Kim de Riel, J., and Liu-Chen, L. Y. (2005) The seventh transmembrane domains of the delta and kappa opioid receptors have different accessibility patterns and interhelical interactions, Biochemistry 44, 16014-16025. 157. Xu, W., Campillo, M., Pardo, L., de Riel, J. K., and Liu-Chen, L.-Y. (2005) The Seventh Transmembrane Domains of the δ and κ Opioid Receptors Have Different Accessibility Patterns and Interhelical Interactions, Biochemistry 44, 16014-16025. 158. Xu, W., Li, J., Chen, C., Huang, P., Weinstein, H., Javitch, J. A., Shi, L., de Riel, J. K., and Liu-Chen, L.-Y. (2001) Comparison of the Amino Acid Residues in the Sixth Transmembrane Domains Accessible in the Binding-Site Crevices of μ, δ, and κ Opioid Receptors, Biochemistry 40, 8018-8029. 159. Chavkin, C., Sud, S., Jin, W., Stewart, J., Zjawiony, J. K., Siebert, D. J., Toth, B. A., Hufeisen, S. J., and Roth, B. L. (2004) Salvinorin A, an active component of the hallucinogenic sage salvia divinorum is a highly efficacious kappa-opioid receptor agonist: structural and functional considerations, J Pharmacol Exp Ther 308, 1197-1203. 160. Vortherms, T. A., Mosier, P. D., Westkaemper, R. B., and Roth, B. L. (2007) Differential helical orientations among related G protein-coupled receptors provide a novel mechanism for selectivity. Studies with salvinorin A and the kappa-opioid receptor, J Biol Chem 282, 3146-3156.

134

161. Baldwin, J. M., Schertler, G. F., and Unger, V. M. (1997) An alpha-carbon template for the transmembrane helices in the rhodopsin family of G-protein- coupled receptors, J Mol Biol 272, 144-164. 162. Palczewski, K., Kumasaka, T., Hori, T., Behnke, C. A., Motoshima, H., Fox, B. A., Le Trong, I., Teller, D. C., Okada, T., Stenkamp, R. E., Yamamoto, M., and Miyano, M. (2000) Crystal structure of rhodopsin: A G protein-coupled receptor [see comments], Science 289, 739-745. 163. Bower, M. J., Cohen, F. E., and Dunbrack, R. L., Jr. (1997) Prediction of protein side-chain rotamers from a backbone-dependent rotamer library: a new homology modeling tool, J Mol Biol 267, 1268-1282. 164. Ronsisvalle, G., Pasquinucci, L., Pappalardo, M. S., Vittorio, F., Fronza, G., Romagnoli, C., Pistacchio, E., Spampinato, S., and Ferri, S. (1993) Non-peptide ligands for opioid receptors. Design of kappa-specific agonists, J Med Chem 36, 1860-1865. 165. Lavecchia, A., Greco, G., Novellino, E., Vittorio, F., and Ronsisvalle, G. (2000) Modeling of kappa-opioid receptor/agonists interactions using pharmacophore- based and docking simulations, J Med Chem 43, 2124-2134. 166. Jones, G., and Willett, P. (1995) Docking small-molecule ligands into active sites, Curr Opin Biotechnol 6, 652-656. 167. Jones, G., Willett, P., Glen, R. C., Leach, A. R., and Taylor, R. (1997) Development and validation of a genetic algorithm for flexible docking, J Mol Biol 267, 727-748. 168. Vortherms, T. A., Mosier, P. D., Westkaemper, R. B., and Roth, B. L. (2007) Differential Helical Orientations among Related G Protein-coupled Receptors Provide a Novel Mechanism for Selectivity: Studies With Salvinorin A and the κ- Opioid Receptor, J. Biol. Chem. 282, 3146-3156. 169. Cantescu, A. A., Shelenkov, A. A., and Dunbrack Jr., R. L. (2003) A Graph- theory Algorithm for Rapid Protein Side-chain Prediction, Protein Sci. 12, 2001- 2014. 170. Karas, M., and Hillenkamp, F. (1988) Laser desorption ionization of proteins with molecular masses exceeding 10,000 daltons, Anal Chem 60, 2299-2301. 171. Ballesteros, J., and Palczewski, K. (2001) G protein-coupled receptor : implications from the crystal structure of rhodopsin, Curr Opin Drug Discov Devel 4, 561-574. 172. Shi, L., and Javitch, J. A. (2004) The second extracellular loop of the dopamine D2 receptor lines the binding-site crevice, Proc Natl Acad Sci U S A 101, 440- 445. 173. Shapiro, D. A., Kristiansen, K., Kroeze, W. K., and Roth, B. L. (2000) Differential modes of agonist binding to 5-hydroxytryptamine(2A) serotonin receptors revealed by mutation and molecular modeling of conserved residues in transmembrane region 5, Mol Pharmacol 58, 877-886. 174. Kristiansen, K. (2004) Molecular mechanisms of ligand binding, signaling, and regulation within the superfamily of G-protein-coupled receptors: molecular modeling and mutagenesis approaches to receptor structure and function, Pharmacol Ther 103, 21-80.

135

175. Xu, W., Li, J., Chen, C., Huang, P., Weinstein, H., Javitch, J. A., Shi, L., de Riel, J. K., and Liu-Chen, L. Y. (2001) Comparison of the amino acid residues in the sixth transmembrane domains accessible in the binding-site crevices of mu, delta, and kappa opioid receptors, Biochemistry 40, 8018-8029. 176. Xu, W., Chen, C., Huang, P., Li, J., de Riel, J. K., Javitch, J. A., and Liu-Chen, L. Y. (2000) The conserved cysteine 7.38 residue is differentially accessible in the binding-site crevices of the mu, delta, and kappa opioid receptors, Biochemistry 39, 13904-13915. 177. Kong, H., Raynor, K., Yano, H., Takeda, J., Bell, G. I., and Reisine, T. (1994) Agonists and antagonists bind to different domains of the cloned kappa opioid receptor, Proc Natl Acad Sci U S A 91, 8042-8046. 178. Ott, D., Frischknecht, R., and Pluckthun, A. (2004) Construction and characterization of a kappa opioid receptor devoid of all free , Protein Eng Des Sel 17, 37-48. 179. Lee, D. Y., Karnati, V. V., He, M., Liu-Chen, L. Y., Kondaveti, L., Ma, Z., Wang, Y., Chen, Y., Beguin, C., Carlezon, W. A., Jr., and Cohen, B. (2005) Synthesis and in vitro pharmacological studies of new C(2) modified salvinorin A analogues, Bioorg Med Chem Lett 15, 3744-3747. 180. Lee, D. Y., He, M., Kondaveti, L., Liu-Chen, L. Y., Ma, Z., Wang, Y., Chen, Y., Li, J. G., Beguin, C., Carlezon, W. A., Jr., and Cohen, B. (2005) Synthesis and in vitro pharmacological studies of C(4) modified salvinorin A analogues, Bioorg Med Chem Lett 15, 4169-4173. 181. Pogozheva, I. D., Lomize, A. L., and Mosberg, H. I. (1998) Opioid receptor three- dimensional structures from distance geometry calculations with hydrogen bonding constraints, Biophys J 75, 612-634. 182. Rasmussen, S. G., Choi, H. J., Rosenbaum, D. M., Kobilka, T. S., Thian, F. S., Edwards, P. C., Burghammer, M., Ratnala, V. R., Sanishvili, R., Fischetti, R. F., Schertler, G. F., Weis, W. I., and Kobilka, B. K. (2007) Crystal structure of the human beta2 adrenergic G-protein-coupled receptor, Nature 450, 383-387. 183. Okada, T., Sugihara, M., Bondar, A.-N., Elstner, M., Entel, P., and Buss, V. (2004) The Retinal Conformation and its Environment in Rhodopsin in Light of a New 2.2 Å Crystal Structure, J. Mol. Biol. 342, 571-583. 184. Devanathan, S., Yao, Z., Salamon, Z., Kobilka, B., and G., T. (2004) Plasmon- Waveguide Resonance Studies of Ligand Binding to the Human β2-Adrenergic Receptor, Biochemistry 43, 3280-3288. 185. Galandrin, S., and Bouvier, M. (2006) Distinct Signaling Profiles of β1 and β2 Adrenergic Receptor Ligands toward Adenylyl Cyclase and Mitogen-Activated Protein Kinase Reveals the Pluridimensionality of Efficacy, Mol. Pharmacol. 70, 1575-1584. 186. Alousi, A., Jasper, J. R., Insel, P. A., and Motulsky, H. J. (1991) Stoichiometry of Receptor-Gs-adenylate Cyclase Interactions, FASEB J. 5, 2300-2303. 187. Nobles, M., Benians, A., and Tinker, A. (2005) Heterotrimeric G Proteins Precouple With G Protein-Coupled Receptors in Living Cells, Proc. Nat. Acad. Sci. USA 102, 18706-18711.

136

188. Kenakin, T. (1997) Differences Between Natural and Recombinant G Protein- coupled Receptor Systems with Varying Receptor/G Protein Stoichiometry, Trends Pharmacol. Sci. 18, 456-464. 189. Connor, M., and Christie, M. J. (1999) Opioid Receptor Signalling Mechanisms, Clin. Exp. Pharmacol. Physiol. 26, 493-499. 190. Lee, J. W. M., Joshi, S., Chan, J. S. C., and Wong, Y. H. (1998) Differential Coupling of μ-, δ-, and κ-Opioid Receptors to Gα16-Mediated Stimulation of Phospholipase C, J. Neurochem. 70, 2203-2211. 191. Berg, K. A., Maayani, S., Goldfarb, J., Scaramellini, C., Leff, P., and Clarke, W. P. (1998) Effector Pathway-Dependent Relative Efficacy at Serotonin Type 2A and 2C Receptors: Evidence for Agonist-Directed Trafficking of Receptor Stimulus, Mol. Pharmacol. 54, 94-104. 192. Gettys, T. W., Fields, T. A., and Raymond, J. R. (1994) Selective Activation of Inhibitory G-Protein α-Subunits by Partial Agonists of the Human 5-HT1A Receptor, Biochemistry 33, 4283-4290. 193. Gray, J. A., Sheffler, D. J., Bhatnagar, A., Woods, J. A., Hufeisen, S. J., Benovic, J. L., and Roth, B. L. (2001) Cell-Type Specific Effects of Endocytosis Inhibitors on 5-Hydroxytryptamine2A Receptor Desensitization and Resensitization Reveal an Arrestin-, GRK2-, and GRK5-Independent Mode of Regulation in Human Embryonic Kidney 293 Cells, Mol. Pharmacol. 60, 1020-1030. 194. Raymond, J. R., Olsen, C. L., and Gettys, T. W. (1993) Cell-Specific Physical and Functional Coupling of Human 5-HT1A Receptors to Inhibitory G Protein α- Subunits and Lack of Coupling to Gsα, Biochemistry 32, 11064-11073. 195. Roth, B. L., and Chuang, D.-M. (1987) Multiple Mechanisms of Signal Transduction, Life Sci. 41, 1051-1064. 196. Aronin, N., and DiFiglia, M. (1992) The Subcellular Localization of the G-Protein Giα in the Reveals Its Potential Role in Both Signal Transduction and Vesicle Trafficking, J. Neurosci. 12, 3435-3444. 197. Giesberts, A. N., van Ginneken, M., Gorter, G., Lapetina, E. G., Akkerman, J.-W. N., and van Willigen, G. (1997) Subcellular Localization of α-Subunits of Trimeric G-Proteins in Human Platelets, Biochem. Biophys. Res. Commun. 234, 439-444. 198. Kawai, Y., and Arinze, I. J. (1996) Differential Localization and Development- dependent Expression of G-protein Subunits, Goα and Gβ, in Rabbit , J. Mol. Cell. Cardiol. 28, 1555-1564. 199. Eglen, R. M. (2005) Emerging Concepts in GPCR Function - The Influence of Cell Phenotype on GPCR Pharmacology, Proc. West. Pharmacol. Soc. 48, 31-34. 200. Krumins, A. M., and Gilman, A. G. (2006) Targeted Knockdown of G Protein Subunits Selectively Prevents Receptor-mediated Modulation of Effectors and Reveals Complex Changes in Non-targeted Signaling Proteins, J. Biol. Chem. 281, 10250-10262. 201. Tinker, A. (2006) The Selective Interactions and Functions of Regulators of G- protein Signalling, Sem. Cell Devel. Biol. 17, 377-382. 202. Rebois, R. V., Robitaille, M., Galés, C., Dupré, D. J., Baragli, A., Trieu, P., Ethier, N., Bouvier, M., and Hérbert, T. E. (2006) Heterotrimeric G Proteins form

137

Stable Complexes with Adenylyl Cyclase and Kir3.1 Channels in Living Cells, J. Cell Sci. 119, 2807-2818. 203. Urban, J. D., Clarke, W. P., von Zastrow, M., Nichols, D. E., Kobilka, B., Weinstein, H., Javitch, J. A., Roth, B. L., Christopoulos, A., Sexton, P. M., Miller, K. J., Spedding, M., and Mailman, R. B. (2007) Functional Selectivity and Classical Concepts of Quantitative Pharmacology, J. Pharmacol. Exp. Ther. 320, 1-13. 204. Farrens, D. L., Altenbach, C., Yang, K., Hubbell, W. L., and Khorana, H. G. (1996) Requirement of rigid-body motion of transmembrane helices for light activation of rhodopsin, Science 274, 768-770. 205. Yao, X., Parnot, C., Deupi, X., Ratnala, V. R., Swaminath, G., Farrens, D., and Kobilka, B. (2006) Coupling ligand structure to specific conformational switches in the beta2-adrenoceptor, Nat Chem Biol 2, 417-422. 206. Javitch, J. A., Fu, D., Liapakis, G., and Chen, J. (1997) Constitutive Activation of the β2 Adrenergic Receptor Alters the Orientation of Its Sixth Membrane- spanning Segment, J. Biol. Chem. 272, 18546-18549. 207. Elling, C. E., Frimurer, T. M., Gerlach, L. O., Jorgensen, R., Holst, B., and Schwartz, T. W. (2006) Metal ion site engineering indicates a global toggle switch model for seven-transmembrane receptor activation, J Biol Chem 281, 17337-17346. 208. Li, J. H., Han, S. J., Hamdan, F. F., Kim, S. K., Jacobson, K. A., Bloodworth, L. M., Zhang, X., and Wess, J. (2007) Distinct structural changes in a G protein- coupled receptor caused by different classes of agonist ligands, J Biol Chem 282, 26284-26293. 209. Gupta, A., Decaillot, F. M., Gomes, I., Tkalych, O., Heimann, A. S., Ferro, E. S., and Devi, L. A. (2007) Conformation state-sensitive antibodies to G-protein- coupled receptors, J Biol Chem 282, 5116-5124. 210. Klein-Seetharaman, J., Yanamala, N. V., Javeed, F., Reeves, P. J., Getmanova, E. V., Loewen, M. C., Schwalbe, H., and Khorana, H. G. (2004) Differential dynamics in the G protein-coupled receptor rhodopsin revealed by solution NMR, Proc Natl Acad Sci U S A 101, 3409-3413. 211. Xu, W., Chen, C., Huang, P., Li, J., de Riel, J. K., Javitch, J. A., and Liu-Chen, L.-Y. (2000) The Conserved Cysteine 7.38 Residue Is Differentially Accessible in the Binding-Site Crevices of the μ, δ, and κ Opioid Receptors, Biochemistry 39, 13904-13915. 212. Kane, B. E., Nieto, M. J., McCurdy, C. R., and Ferguson, D. M. (2006) A Unique Binding Epitope for Salvinorin A, a Non-nitrogenous Kappa Opioid Receptor Agonist, FEBS J. 273, 1966-1974. 213. Buck, E., and Wells, J. A. (2005) Disulfide Trapping to Localize Small-molecule Agonists and Antagonists for a G Protein-coupled Receptor, Proc. Nat. Acad. Sci. USA 102, 2719-2724. 214. Elling, C. E., Frimurer, T. M., Gerlach, L.-O., Jorgensen, R., Holst, B., and Schwartz, T. W. (2006) Metal Ion Site Engineering Indicates a Global Toggle Switch Model for Seven-transmembrane Receptor Activation, J. Biol. Chem. 281, 17337-17346.

138

215. Han, S.-J., Hamdan, F. F., Kim, S.-K., Jacobsen, K. A., Bloodworth, L. M., Li, B., and Wess, J. (2005) Identification of an Agonist-induced Conformational Change Occurring Adjacent to the Ligand-binding Pocket of the M3 Muscarinic Receptor, J. Biol. Chem. 280, 34849-34858. 216. Galés, C., Rebois, R. V., Hogue, M., Trieu, P., Breit, A., Hérbert, T. E., and Bouvier, M. (2005) Real-time Monitoring of Receptor and G-protein Interactions in Living Cells, Nature Methods 2, 177-184. 217. Neubig, R. R., Gantzos, R. D., and Thompsen, W. J. (1988) Mechanism of Agonist and Antagonist Binding to α2 Adrenergic Receptors: Evidence for a Precoupled Receptor-Guanine Nucleotide Protein Complex, Biochemistry 27, 2374-2384. 218. Holmgren, M., Liu, Y., Xu, Y., and Yellen, G. (1996) On the Use of Thiol- modifying Agents to Determine Channel Topology, Neuropharmacology 35, 797- 804. 219. Shi, L., and Javitch, J. A. (2004) The Second Extracellular Loop of the Dopamine D2 Receptor Lines the Binding-site Crevice, Proc. Nat. Acad. Sci. USA 101, 440- 445. 220. Harding, W. W., Schmidt, M., Tidgewell, K., Kannan, P., Holden, K. G., Dersch, C. M., Rothman, R. B., and Prisinzano, T. E. (2006) Synthetic studies of neoclerodane diterpenes from Salvia divinorum: selective modification of the furan ring, Bioorg Med Chem Lett 16, 3170-3174. 221. Harding, W. W., Tidgewell, K., Schmidt, M., Shah, K., Dersch, C. M., Snyder, J., Parrish, D., Deschamps, J. R., Rothman, R. B., and Prisinzano, T. E. (2005) Salvinicins A and B, new neoclerodane diterpenes from Salvia divinorum, Org Lett 7, 3017-3020. 222. Harding, W. W., Tidgewell, K., Byrd, N., Cobb, H., Dersch, C. M., Butelman, E. R., Rothman, R. B., and Prisinzano, T. E. (2005) Neoclerodane diterpenes as a novel scaffold for mu opioid receptor ligands, J Med Chem 48, 4765-4771. 223. Rothman, R. B., Murphy, D. L., Xu, H., Godin, J. A., Dersch, C. M., Partilla, J. S., Tidgewell, K., Schmidt, M., and Prisinzano, T. E. (2007) Salvinorin A: allosteric interactions at the mu-opioid receptor, J Pharmacol Exp Ther 320, 801- 810. 224. Groer, C. E., Tidgewell, K., Moyer, R. A., Harding, W. W., Rothman, R. B., Prisinzano, T. E., and Bohn, L. M. (2007) An opioid agonist that does not induce micro-opioid receptor--arrestin interactions or receptor internalization, Mol Pharmacol 71, 549-557. 225. Simpson, D. S., Katavic, P. L., Lozama, A., Harding, W. W., Parrish, D., Deschamps, J. R., Dersch, C. M., Partilla, J. S., Rothman, R. B., Navarro, H., and Prisinzano, T. E. (2007) Synthetic studies of neoclerodane diterpenes from Salvia divinorum: preparation and opioid receptor activity of salvinicin analogues, J Med Chem 50, 3596-3603. 226. Tan, K. R., Gonthier, A., Baur, R., Ernst, M., Goeldner, M., and Sigel, E. (2007) Proximity-accelerated chemical coupling reaction in the benzodiazepine-binding site of gamma-aminobutyric acid type A receptors: superposition of different allosteric modulators, J Biol Chem 282, 26316-26325.

139

227. Luo, Y., Knuckley, B., Bhatia, M., Pellechia, P. J., and Thompson, P. R. (2006) Activity-based protein profiling reagents for protein arginine deiminase 4 (PAD4): synthesis and in vitro evaluation of a fluorescently labeled probe, J Am Chem Soc 128, 14468-14469. 228. Cohen, M. S., Zhang, C., Shokat, K. M., and Taunton, J. (2005) Structural bioinformatics-based design of selective, irreversible kinase inhibitors, Science 308, 1318-1321. 229. Cohen, M. S., Hadjivassiliou, H., and Taunton, J. (2007) A clickable inhibitor reveals context-dependent autoactivation of p90 RSK, Nat Chem Biol 3, 156-160. 230. Mattras, H., Aliau, S., Richard, E., Bonnafous, J. C., Jouin, P., and Borgna, J. L. (2002) Identification by MALDI-TOF mass spectrometry of 17 alpha- bromoacetamidopropylestradiol covalent attachment sites on estrogen receptor alpha, Biochemistry 41, 15713-15727. 231. Mattras, H., Aliau, S., Demey, E., Poncet, J., and Borgna, J. L. (2006) Mass spectrometry identification of covalent attachment sites of two related estrogenic ligands on human estrogen receptor alpha, J Biochem Mol Biol 98, 236- 247. 232. Lobaccaro, C., Pons, J. F., Duchesne, M. J., Auzou, G., Pons, M., Nique, F., Teutsch, G., and Borgna, J. L. (1997) Steroidal affinity labels of the estrogen receptor. 3. 11 beta-n-alkyl derivatives bearing a terminal electrophilic group: antiestrogenic and cytotoxic properties, J Med Chem 40, 2217-2227. 233. Aliau, S., Mattras, H., and Borgna, J. L. (2006) Identification of covalent attachment site of antiestrogenic estradiol 11 beta-derivatives on human estrogen receptor alpha ligand-binding domain, J Steroid Biochem Mol Biol 98, 111-121. 234. Aliau, S., El Garrouj, D., Yasri, A., Katzenellenbogen, B. S., and Borgna, J. L. (1997) 17 alpha (haloacetamidoalkyl) estradiols alkylate the human estrogen receptor at cysteine residues 417 and 530, Biochemistry 36, 5861-5867. 235. Aliau, S., Delettre, G., Mattras, H., El Garrouj, D., Nique, F., Teutsch, G., and Borgna, J. L. (2000) Steroidal affinity labels of the estrogen receptor alpha. 4. Electrophilic 11beta-aryl derivatives of estradiol, J Med Chem 43, 613-628. 236. Foucaud, B., Laube, B., Schemm, R., Kreimeyer, A., Goeldner, M., and Betz, H. (2003) Structural model of the N-methyl-D-aspartate receptor glycine site probed by site-directed chemical coupling, J Biol Chem 278, 24011-24017. 237. Wannemacher, K. M., Yadav, P. N., and Howells, R. D. (2007) A select set of opioid ligands induce up-regulation by promoting the maturation and stability of the rat kappa-opioid receptor in human embryonic kidney 293 cells, J Pharmacol Exp Ther 323, 614-625. 238. Ouagazzal, A. M., Jenck, F., and Moreau, J. L. (2001) Drug-induced potentiation of prepulse inhibition of acoustic startle reflex in mice: a model for detecting antipsychotic activity?, Psychopharmacology (Berl) 156, 273-283. 239. Geyer, M. A., McIlwain, K. L., and Paylor, R. (2002) Mouse genetic models for prepulse inhibition: an early review, Mol 7, 1039-1053. 240. Geyer, M. A., Krebs-Thomson, K., Braff, D. L., and Swerdlow, N. R. (2001) Pharmacological studies of prepulse inhibition models of sensorimotor gating deficits in schizophrenia: a decade in review, Psychopharmacology (Berl) 156, 117-154.

140

241. Braff, D. L., Geyer, M. A., and Swerdlow, N. R. (2001) Human studies of prepulse inhibition of startle: normal subjects, patient groups, and pharmacological studies, Psychopharmacology (Berl) 156, 234-258. 242. Zvonok, N., Yaddanapudi, S., Williams, J., Dai, S., Dong, K., Rejtar, T., Karger, B. L., and Makriyannis, A. (2007) Comprehensive proteomic mass spectrometric characterization of human cannabinoid CB2 receptor, J Proteome Res 6, 2068- 2079. 243. Xu, W., Filppula, S. A., Mercier, R., Yaddanapudi, S., Pavlopoulos, S., Cai, J., Pierce, W. M., and Makriyannis, A. (2005) Purification and mass spectroscopic analysis of human CB1 cannabinoid receptor functionally expressed using the baculovirus system, J Pept Res 66, 138-150. 244. Christoffers, K. H., Li, H., Keenan, S. M., and Howells, R. D. (2003) Purification and mass spectrometric analysis of the mu opioid receptor, Brain Res Mol Brain Res 118, 119-131. 245. Christoffers, K. H., Li, H., and Howells, R. D. (2005) Purification and mass spectrometric analysis of the delta opioid receptor, Brain Res Mol Brain Res 136, 54-64. 246. Christoffers, K. H., Khokhar, A., Chaturvedi, K., and Howells, R. D. (2002) Inhibition of mu and delta opioid receptor ligand binding by the peptide aldehyde protease inhibitor, leupeptin, Regul Pept 105, 9-14. 247. Granvogl, B., Ploscher, M., and Eichacker, L. A. (2007) Sample preparation by in-gel digestion for mass spectrometry-based proteomics, Anal Bioanal Chem 389, 991-1002. 248. Wu, S. L., Kim, J., Hancock, W. S., and Karger, B. (2005) Extended Range Proteomic Analysis (ERPA): a new and sensitive LC-MS platform for high sequence coverage of complex proteins with extensive post-translational modifications-comprehensive analysis of beta-casein and epidermal growth factor receptor (EGFR), J Proteome Res 4, 1155-1170. 249. Sansuk, K., Balog, C. I., van der Does, A. M., Booth, R., de Grip, W. J., Deelder, A. M., Bakker, R. A., Leurs, R., and Hensbergen, P. J. (2008) GPCR Proteomics: Mass Spectrometric and Functional Analysis of Histamine H1 Receptor after Baculovirus-Driven and in Vitro Cell Free Expression, J Proteome Res. 250. Ho, J. T., White, J. F., Grisshammer, R., and Hess, S. (2007) Analysis of a G protein-coupled receptor for neurotensin by liquid chromatography--mass spectrometry, Anal Biochem. 251. Kraft, P., Mills, J., and Dratz, E. (2001) Mass spectrometric analysis of cyanogen bromide fragments of integral membrane proteins at the picomole level: application to rhodopsin, Anal Biochem 292, 76-86. 252. Aebersold, R., and Mann, M. (2003) Mass spectrometry-based proteomics, Nature 422, 198-207. 253. Pan, S., and Aebersold, R. (2007) Quantitative proteomics by stable isotope labeling and mass spectrometry, Methods Mol Biol 367, 209-218. 254. Andersen, J. S., Lam, Y. W., Leung, A. K., Ong, S. E., Lyon, C. E., Lamond, A. I., and Mann, M. (2005) Nucleolar proteome dynamics, Nature 433, 77-83.

141

255. Blagoev, B., Kratchmarova, I., Ong, S. E., Nielsen, M., Foster, L. J., and Mann, M. (2003) A proteomics strategy to elucidate functional protein-protein interactions applied to EGF signaling, Nat Biotechnol 21, 315-318. 256. Chalmers, M. J., Busby, S. A., Pascal, B. D., He, Y., Hendrickson, C. L., Marshall, A. G., and Griffin, P. R. (2006) Probing protein ligand interactions by automated hydrogen/deuterium exchange mass spectrometry, Anal Chem 78, 1005-1014. 257. Wales, T. E., and Engen, J. R. (2006) Hydrogen exchange mass spectrometry for the analysis of protein dynamics, Mass Spectrom Rev 25, 158-170. 258. Li, J., Lim, M. S., Li, S., Brock, M., Pique, M. E., Woods, V. L., Jr., and Craig, L. (2008) Vibrio cholerae toxin-coregulated pilus structure analyzed by hydrogen/deuterium exchange mass spectrometry, Structure 16, 137-148. 259. Kobilka, B., and Schertler, G. F. (2008) New G-protein-coupled receptor crystal structures: insights and limitations, Trends Pharmacol Sci.

142